Concepts of mental life across cultures: Fit between individuals and cultural modesl
package ‘betareg’ was built under R version 4.0.2In this notebook we conduct exploratory factor analyses (EFAs) on the datasets for our studies of concepts of mental life, in which each participants judged the various mental capacities of a particular target entity. We analyze datasets for adults and children from each of our five field sites: the Ghana, Ghana, Thailand, China, and Vanuatu.
This notebook contains an exploration of how well the cultural model represented by the EFA solution describes the responses of individuals within that culture, and whether this “fit” between individtual and cultural model varies along demographic lines.
NOTE: As of now, the “efa_oblique.Rmd” notebook (or one of the alternative versions of these analyses) must be run prior to this notebook.
Functions
US
US adults
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_us_adults)
Residuals:
Min 1Q Median 3Q Max
-1.5642 -0.9119 -0.2456 0.9984 1.8021
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.0002467 0.0897358 -0.003 0.998
scale(age) 0.1300386 0.0901028 1.443 0.152
Residual standard error: 0.9952 on 121 degrees of freedom
(4 observations deleted due to missingness)
Multiple R-squared: 0.01692, Adjusted R-squared: 0.008798
F-statistic: 2.083 on 1 and 121 DF, p-value: 0.1515
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_us_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.7855 -0.8601 0.0118 0.9664 1.4624
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.12880 0.07582 -14.888 <2e-16 ***
scale(age) 0.11029 0.07085 1.557 0.12
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 6.576 0.807 8.149 3.67e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 67.82 on 3 Df
Pseudo R-squared: 0.02108
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_us_adults)
Residuals:
Min 1Q Median 3Q Max
-1.5915 -0.8355 -0.1464 0.8871 1.7750
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.02207 0.08815 -0.25 0.803
gender_m 0.18686 0.08815 2.12 0.036 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9864 on 125 degrees of freedom
Multiple R-squared: 0.0347, Adjusted R-squared: 0.02698
F-statistic: 4.494 on 1 and 125 DF, p-value: 0.036
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_us_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.7742 -0.8282 0.1007 0.8920 1.4238
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.14404 0.07537 -15.179 <2e-16 ***
gender_m 0.12420 0.07124 1.743 0.0813 .
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 6.6103 0.7985 8.278 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 70.29 on 3 Df
Pseudo R-squared: 0.02485
Number of iterations: 13 (BFGS) + 2 (Fisher scoring) Race/ethnicity
Call:
lm(formula = scale(MSE) ~ ethnicity_cat2, data = d_sim_us_adults)
Residuals:
Min 1Q Median 3Q Max
-1.4867 -0.9983 -0.1019 0.9747 1.7861
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.007241 0.092817 0.078 0.938
ethnicity_cat2_POC 0.052715 0.092817 0.568 0.571
Residual standard error: 1.002 on 115 degrees of freedom
(10 observations deleted due to missingness)
Multiple R-squared: 0.002797, Adjusted R-squared: -0.005874
F-statistic: 0.3226 on 1 and 115 DF, p-value: 0.5712
Call:
betareg(formula = MSE_rescaled ~ ethnicity_cat2, data = d_sim_us_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.5746 -0.9836 0.1540 0.9333 1.4319
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.12096 0.07844 -14.290 <2e-16 ***
ethnicity_cat2_POC 0.03417 0.07410 0.461 0.645
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 6.4215 0.8067 7.96 1.72e-15 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 62.87 on 3 Df
Pseudo R-squared: 0.001887
Number of iterations: 13 (BFGS) + 1 (Fisher scoring) Education
Call:
lm(formula = scale(MSE) ~ scale(education_catX), data = d_sim_us_adults %>%
mutate(education_catX = as.numeric(education_catX)))
Residuals:
Min 1Q Median 3Q Max
-1.4282 -1.0083 -0.1788 0.9651 1.7513
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.00410 0.09322 -0.044 0.965
scale(education_catX) -0.03958 0.09361 -0.423 0.673
Residual standard error: 1.013 on 116 degrees of freedom
(9 observations deleted due to missingness)
Multiple R-squared: 0.001539, Adjusted R-squared: -0.007069
F-statistic: 0.1788 on 1 and 116 DF, p-value: 0.6732
Call:
betareg(formula = MSE_rescaled ~ scale(education_catX), data = d_sim_us_adults %>%
mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.4538 -0.9727 0.0934 0.9217 1.4091
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.12998 0.07862 -14.373 <2e-16 ***
scale(education_catX) -0.02931 0.07426 -0.395 0.693
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 6.3292 0.7921 7.991 1.34e-15 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 63.51 on 3 Df
Pseudo R-squared: 0.00126
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) 
Call:
lm(formula = scale(MSE) ~ education_cat2, data = d_sim_us_adults)
Residuals:
Min 1Q Median 3Q Max
-1.3937 -0.9960 -0.1715 0.9799 1.7526
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.04839 0.11068 0.437 0.663
education_cat2_coll -0.09678 0.11068 -0.874 0.384
Residual standard error: 1.01 on 116 degrees of freedom
(9 observations deleted due to missingness)
Multiple R-squared: 0.006549, Adjusted R-squared: -0.002015
F-statistic: 0.7647 on 1 and 116 DF, p-value: 0.3837
Call:
betareg(formula = MSE_rescaled ~ education_cat2, data = d_sim_us_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.3865 -0.9613 0.1003 0.9406 1.4027
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.09348 0.09074 -12.051 <2e-16 ***
education_cat2_coll -0.06831 0.08721 -0.783 0.433
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 6.3544 0.7954 7.989 1.36e-15 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 63.74 on 3 Df
Pseudo R-squared: 0.005235
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) Rural/urban
Call:
lm(formula = scale(MSE) ~ urban_rural_cat2, data = d_sim_us_adults)
Residuals:
Min 1Q Median 3Q Max
-1.4880 -0.9110 -0.1119 0.9740 1.7746
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.03607 0.10621 -0.340 0.735
urban_rural_cat2_rural -0.09736 0.10621 -0.917 0.361
Residual standard error: 0.9975 on 119 degrees of freedom
(6 observations deleted due to missingness)
Multiple R-squared: 0.007011, Adjusted R-squared: -0.001333
F-statistic: 0.8402 on 1 and 119 DF, p-value: 0.3612
Call:
betareg(formula = MSE_rescaled ~ urban_rural_cat2, data = d_sim_us_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.5787 -0.8385 0.1561 0.9412 1.4173
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.14201 0.08924 -12.797 <2e-16 ***
urban_rural_cat2_rural -0.05178 0.08547 -0.606 0.545
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 6.4848 0.8011 8.095 5.74e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 65.17 on 3 Df
Pseudo R-squared: 0.003224
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) Religion
Call:
lm(formula = scale(MSE) ~ religion_cat3, data = d_sim_us_adults)
Residuals:
Min 1Q Median 3Q Max
-1.4481 -1.0537 -0.1498 0.9614 1.7193
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.013403 0.120930 0.111 0.912
religion_cat3_christian -0.003784 0.164282 -0.023 0.982
religion_cat3_other -0.004167 0.203114 -0.021 0.984
Residual standard error: 1.019 on 109 degrees of freedom
(15 observations deleted due to missingness)
Multiple R-squared: 3.228e-05, Adjusted R-squared: -0.01832
F-statistic: 0.001759 on 2 and 109 DF, p-value: 0.9982
Call:
betareg(formula = MSE_rescaled ~ religion_cat3, data = d_sim_us_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.4315 -0.9989 0.0756 0.9348 1.3966
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.10031 0.09887 -11.129 <2e-16 ***
religion_cat3_christian 0.01814 0.12942 0.140 0.889
religion_cat3_other 0.01301 0.16005 0.081 0.935
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 6.2709 0.8045 7.795 6.45e-15 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 59.18 on 4 Df
Pseudo R-squared: 0.0008433
Number of iterations: 15 (BFGS) + 2 (Fisher scoring) Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_us_adults, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-1.5333 -0.4267 -0.1060 0.5325 2.1092
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.004821 0.069308 0.070 0.944662
target1 -0.873472 0.205658 -4.247 4.36e-05 ***
target2 0.278033 0.205658 1.352 0.179009
target3 1.037701 0.213118 4.869 3.54e-06 ***
target4 0.094057 0.205658 0.457 0.648269
target5 -0.111238 0.205658 -0.541 0.589613
target6 -0.660440 0.205658 -3.211 0.001706 **
target7 -1.147288 0.213118 -5.383 3.82e-07 ***
target8 0.721863 0.213118 3.387 0.000963 ***
target9 0.371713 0.205658 1.807 0.073264 .
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.7805 on 117 degrees of freedom
Multiple R-squared: 0.4343, Adjusted R-squared: 0.3908
F-statistic: 9.98 on 9 and 117 DF, p-value: 2.949e-11
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_us_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.3419 -0.5768 -0.0069 0.6897 2.0609
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.19369 0.06117 -19.513 < 2e-16 ***
target1 -0.74992 0.19730 -3.801 0.000144 ***
target2 0.30231 0.16166 1.870 0.061487 .
target3 0.85115 0.15865 5.365 8.10e-08 ***
target4 0.11247 0.16634 0.676 0.498942
target5 -0.10809 0.17287 -0.625 0.531769
target6 -0.42676 0.18414 -2.318 0.020470 *
target7 -1.12955 0.22184 -5.092 3.55e-07 ***
target8 0.64964 0.16091 4.037 5.40e-05 ***
target9 0.29374 0.16185 1.815 0.069547 .
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 11.617 1.434 8.101 5.44e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 104.9 on 11 Df
Pseudo R-squared: 0.4744
Number of iterations: 20 (BFGS) + 2 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + scale(education_catX) +
ethnicity_cat2 + religion_cat3 + urban_rural_cat2 + target,
data = d_sim_us_adults %>% mutate(education_catX = as.numeric(education_catX)))
Residuals:
Min 1Q Median 3Q Max
-1.49646 -0.48825 0.01192 0.52053 1.91945
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.16833 0.11476 -1.467 0.146013
scale(age) 0.09620 0.08764 1.098 0.275393
gender_m 0.18959 0.08575 2.211 0.029663 *
scale(education_catX) -0.02017 0.08699 -0.232 0.817224
ethnicity_cat2_POC 0.09782 0.08527 1.147 0.254465
religion_cat3_christian 0.15160 0.14212 1.067 0.289053
religion_cat3_other -0.20335 0.16734 -1.215 0.227573
urban_rural_cat2_rural -0.14730 0.10039 -1.467 0.145896
target1 -0.88628 0.22850 -3.879 0.000204 ***
target2 0.15175 0.25431 0.597 0.552233
target3 0.96683 0.23034 4.197 6.49e-05 ***
target4 0.13816 0.25225 0.548 0.585274
target5 -0.15699 0.23782 -0.660 0.510931
target6 -0.58680 0.23668 -2.479 0.015093 *
target7 -1.21028 0.24156 -5.010 2.82e-06 ***
target8 0.73389 0.24395 3.008 0.003434 **
target9 0.63267 0.25004 2.530 0.013198 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.7832 on 87 degrees of freedom
(23 observations deleted due to missingness)
Multiple R-squared: 0.502, Adjusted R-squared: 0.4104
F-statistic: 5.481 on 16 and 87 DF, p-value: 6.894e-08
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + education_cat2 +
ethnicity_cat2 + religion_cat3 + urban_rural_cat2 + target,
data = d_sim_us_adults)
Residuals:
Min 1Q Median 3Q Max
-1.49828 -0.48186 0.00609 0.50569 1.95093
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.12682 0.13073 -0.970 0.33470
scale(age) 0.09832 0.08646 1.137 0.25859
gender_m 0.18479 0.08484 2.178 0.03210 *
education_cat2_coll -0.06855 0.09829 -0.697 0.48740
ethnicity_cat2_POC 0.09581 0.08437 1.136 0.25925
religion_cat3_christian 0.15947 0.14214 1.122 0.26499
religion_cat3_other -0.20625 0.16611 -1.242 0.21771
urban_rural_cat2_rural -0.14099 0.09901 -1.424 0.15800
target1 -0.90207 0.22632 -3.986 0.00014 ***
target2 0.13824 0.25151 0.550 0.58397
target3 0.97767 0.22819 4.285 4.71e-05 ***
target4 0.15971 0.25369 0.630 0.53063
target5 -0.16960 0.23755 -0.714 0.47717
target6 -0.58222 0.23608 -2.466 0.01562 *
target7 -1.20051 0.24038 -4.994 3.01e-06 ***
target8 0.71676 0.24266 2.954 0.00404 **
target9 0.63535 0.24946 2.547 0.01263 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.7812 on 87 degrees of freedom
(23 observations deleted due to missingness)
Multiple R-squared: 0.5044, Adjusted R-squared: 0.4133
F-statistic: 5.535 on 16 and 87 DF, p-value: 5.74e-08Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + scale(education_catX) + ethnicity_cat2 +
religion_cat3 + urban_rural_cat2 + +(1 | target)
Data: d_sim_us_adults %>% mutate(education_catX = as.numeric(education_catX))
REML criterion at convergence: 278.9
Scaled residuals:
Min 1Q Median 3Q Max
-1.92364 -0.65918 -0.02763 0.66739 2.31261
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.4546 0.6742
Residual 0.6127 0.7828
Number of obs: 104, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) -0.16244 0.24195 11.62404 -0.671 0.5151
scale(age) 0.10379 0.08716 88.82348 1.191 0.2369
gender_m 0.18457 0.08504 89.68817 2.170 0.0326 *
scale(education_catX) -0.01849 0.08609 90.26392 -0.215 0.8304
ethnicity_cat2_POC 0.09933 0.08474 89.04396 1.172 0.2443
religion_cat3_christian 0.14054 0.14141 88.66858 0.994 0.3230
religion_cat3_other -0.18618 0.16653 88.62031 -1.118 0.2666
urban_rural_cat2_rural -0.14549 0.09983 88.83638 -1.457 0.1485
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Correlation of Fixed Effects:
(Intr) scl(g) gndr_m sc(_X) e_2_PO rlgn_ct3_c rlgn_ct3_t
scale(age) -0.001
gender_m -0.090 -0.038
scl(dctn_X) -0.089 -0.168 0.187
ethnc_2_POC -0.055 0.138 0.070 0.147
rlgn_ct3_ch -0.083 -0.160 0.111 0.017 0.185
rlgn_ct3_th 0.220 0.016 -0.044 -0.107 -0.135 -0.719
urbn_rrl_2_ 0.245 0.065 -0.167 -0.217 -0.056 -0.174 0.115 Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + education_cat2 + ethnicity_cat2 +
religion_cat3 + urban_rural_cat2 + +(1 | target)
Data: d_sim_us_adults
REML criterion at convergence: 278.2
Scaled residuals:
Min 1Q Median 3Q Max
-1.85218 -0.67475 -0.04646 0.66937 2.35719
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.4558 0.6751
Residual 0.6097 0.7809
Number of obs: 104, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) -0.12208 0.25010 13.17602 -0.488 0.6335
scale(age) 0.10595 0.08607 88.51948 1.231 0.2216
gender_m 0.17993 0.08419 89.49226 2.137 0.0353 *
education_cat2_coll -0.06607 0.09751 89.58791 -0.678 0.4998
ethnicity_cat2_POC 0.09721 0.08392 88.78821 1.158 0.2498
religion_cat3_christian 0.14791 0.14140 88.73315 1.046 0.2984
religion_cat3_other -0.18827 0.16519 88.84221 -1.140 0.2575
urban_rural_cat2_rural -0.13926 0.09854 88.58445 -1.413 0.1611
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Correlation of Fixed Effects:
(Intr) scl(g) gndr_m edc_2_ e_2_PO rlgn_ct3_c rlgn_ct3_t
scale(age) 0.009
gender_m -0.109 -0.020
edctn_ct2_c -0.266 -0.092 0.142
ethnc_2_POC -0.063 0.158 0.056 0.084
rlgn_ct3_ch -0.059 -0.152 0.099 -0.071 0.178
rlgn_ct3_th 0.209 -0.001 -0.026 -0.017 -0.122 -0.718
urbn_rrl_2_ 0.263 0.044 -0.152 -0.163 -0.038 -0.160 0.096
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + scale(education_catX) + ethnicity_cat2 +
religion_cat3 + urban_rural_cat2 + target, data = d_sim_us_adults %>% mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.8910 -0.5708 0.0955 0.8445 2.2503
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.34960 0.09628 -14.017 < 2e-16 ***
scale(age) 0.08158 0.06953 1.173 0.240667
gender_m 0.16689 0.06855 2.435 0.014907 *
scale(education_catX) -0.02301 0.06801 -0.338 0.735058
ethnicity_cat2_POC 0.08825 0.06826 1.293 0.196036
religion_cat3_christian 0.16149 0.11216 1.440 0.149916
religion_cat3_other -0.18048 0.13184 -1.369 0.171034
urban_rural_cat2_rural -0.11848 0.08010 -1.479 0.139099
target1 -0.79632 0.20866 -3.816 0.000135 ***
target2 0.22505 0.19332 1.164 0.244363
target3 0.80087 0.16617 4.820 1.44e-06 ***
target4 0.13648 0.19688 0.693 0.488182
target5 -0.18789 0.19251 -0.976 0.329067
target6 -0.38232 0.20265 -1.887 0.059216 .
target7 -1.22749 0.24446 -5.021 5.14e-07 ***
target8 0.69093 0.17866 3.867 0.000110 ***
target9 0.56901 0.18429 3.088 0.002018 **
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 12.837 1.758 7.304 2.8e-13 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 91.99 on 18 Df
Pseudo R-squared: 0.5404
Number of iterations: 26 (BFGS) + 1 (Fisher scoring)
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + education_cat2 + ethnicity_cat2 +
religion_cat3 + urban_rural_cat2 + target, data = d_sim_us_adults %>% mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.7908 -0.5328 0.1031 0.8451 2.2701
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.31252 0.10747 -12.213 < 2e-16 ***
scale(age) 0.08050 0.06891 1.168 0.242731
gender_m 0.16163 0.06797 2.378 0.017413 *
education_cat2_coll -0.06522 0.07757 -0.841 0.400519
ethnicity_cat2_POC 0.08660 0.06753 1.282 0.199719
religion_cat3_christian 0.16854 0.11217 1.502 0.132973
religion_cat3_other -0.18544 0.13107 -1.415 0.157119
urban_rural_cat2_rural -0.11362 0.07899 -1.438 0.150328
target1 -0.81368 0.20741 -3.923 8.74e-05 ***
target2 0.21952 0.19106 1.149 0.250560
target3 0.81165 0.16460 4.931 8.18e-07 ***
target4 0.15808 0.19825 0.797 0.425246
target5 -0.19078 0.19201 -0.994 0.320421
target6 -0.38137 0.20226 -1.886 0.059354 .
target7 -1.22328 0.24382 -5.017 5.25e-07 ***
target8 0.67580 0.17805 3.796 0.000147 ***
target9 0.56767 0.18396 3.086 0.002030 **
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 12.915 1.769 7.303 2.82e-13 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 92.29 on 18 Df
Pseudo R-squared: 0.5429
Number of iterations: 27 (BFGS) + 1 (Fisher scoring) US children
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_us_children)
Residuals:
Min 1Q Median 3Q Max
-1.64038 -0.78859 -0.09887 0.63780 2.35655
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.02249 0.09002 0.250 0.8032
scale(age) -0.22944 0.09043 -2.537 0.0126 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9484 on 109 degrees of freedom
(6 observations deleted due to missingness)
Multiple R-squared: 0.05577, Adjusted R-squared: 0.04711
F-statistic: 6.438 on 1 and 109 DF, p-value: 0.01259
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_us_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.2633 -0.8050 0.0276 0.7177 2.1586
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.82621 0.05779 -14.298 <2e-16 ***
scale(age) -0.14536 0.05725 -2.539 0.0111 *
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 11.743 1.525 7.701 1.35e-14 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 73.48 on 3 Df
Pseudo R-squared: 0.05834
Number of iterations: 14 (BFGS) + 2 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_us_children)
Residuals:
Min 1Q Median 3Q Max
-1.60311 -0.82274 -0.08055 0.56064 2.28751
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.006531 0.093753 -0.070 0.945
gender_m -0.044949 0.093753 -0.479 0.633
Residual standard error: 1.003 on 115 degrees of freedom
Multiple R-squared: 0.001995, Adjusted R-squared: -0.006683
F-statistic: 0.2299 on 1 and 115 DF, p-value: 0.6325
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_us_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.1166 -0.7964 0.0264 0.6006 2.0272
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.83884 0.06017 -13.940 <2e-16 ***
gender_m -0.01268 0.05907 -0.215 0.83
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 10.378 1.308 7.937 2.07e-15 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 71.37 on 3 Df
Pseudo R-squared: 0.0004117
Number of iterations: 11 (BFGS) + 2 (Fisher scoring) Race/ethnicity
Call:
lm(formula = scale(MSE) ~ ethnicity_cat2, data = d_sim_us_children)
Residuals:
Min 1Q Median 3Q Max
-1.64875 -0.72877 -0.05011 0.49631 2.30261
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.02133 0.10457 -0.204 0.839
ethnicity_cat2_POC -0.16071 0.10457 -1.537 0.128
Residual standard error: 0.9403 on 90 degrees of freedom
(25 observations deleted due to missingness)
Multiple R-squared: 0.02557, Adjusted R-squared: 0.01475
F-statistic: 2.362 on 1 and 90 DF, p-value: 0.1278
Call:
betareg(formula = MSE_rescaled ~ ethnicity_cat2, data = d_sim_us_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.2593 -0.6960 0.0715 0.5952 2.1006
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.86005 0.06807 -12.634 <2e-16 ***
ethnicity_cat2_POC -0.09038 0.06701 -1.349 0.177
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 11.56 1.65 7.006 2.45e-12 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 61.7 on 3 Df
Pseudo R-squared: 0.01968
Number of iterations: 13 (BFGS) + 2 (Fisher scoring) Religion
Call:
lm(formula = scale(MSE) ~ religion_cat3, data = d_sim_us_children)
Residuals:
Min 1Q Median 3Q Max
-1.44597 -0.74331 -0.09855 0.42893 2.02047
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.07873 0.10215 -0.771 0.4427
religion_cat3_christian 0.18832 0.13128 1.435 0.1546
religion_cat3_other -0.36717 0.15798 -2.324 0.0222 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.979 on 98 degrees of freedom
(16 observations deleted due to missingness)
Multiple R-squared: 0.05282, Adjusted R-squared: 0.03349
F-statistic: 2.732 on 2 and 98 DF, p-value: 0.07002
Call:
betareg(formula = MSE_rescaled ~ religion_cat3, data = d_sim_us_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.7941 -0.6858 0.0282 0.5214 1.8772
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.89102 0.06637 -13.425 <2e-16 ***
religion_cat3_christian 0.12197 0.08284 1.472 0.1409
religion_cat3_other -0.23630 0.10274 -2.300 0.0215 *
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 11.136 1.514 7.354 1.93e-13 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 65.23 on 4 Df
Pseudo R-squared: 0.05758
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_us_children, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-1.4564 -0.6476 -0.1414 0.6052 2.2678
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.01645 0.07969 -0.206 0.836891
target1 -0.35622 0.24507 -1.454 0.149006
target2 0.16290 0.23577 0.691 0.491109
target3 -0.04336 0.22759 -0.191 0.849259
target4 -0.31437 0.23577 -1.333 0.185234
target5 -0.32376 0.22759 -1.423 0.157780
target6 -0.39859 0.24507 -1.626 0.106802
target7 -0.85152 0.25580 -3.329 0.001197 **
target8 0.16653 0.24507 0.680 0.498275
target9 0.84887 0.23577 3.600 0.000483 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8594 on 107 degrees of freedom
Multiple R-squared: 0.3188, Adjusted R-squared: 0.2615
F-statistic: 5.563 on 9 and 107 DF, p-value: 2.82e-06
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_us_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.0793 -0.7722 -0.0763 0.7537 2.4489
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.87368 0.05131 -17.027 < 2e-16 ***
target1 -0.20200 0.15840 -1.275 0.202233
target2 0.14060 0.14445 0.973 0.330387
target3 -0.04654 0.14326 -0.325 0.745286
target4 -0.22590 0.15302 -1.476 0.139885
target5 -0.17757 0.14642 -1.213 0.225229
target6 -0.24992 0.15981 -1.564 0.117866
target7 -0.60272 0.17970 -3.354 0.000796 ***
target8 0.12712 0.15037 0.845 0.397898
target9 0.54206 0.13908 3.898 9.72e-05 ***
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 15.255 1.946 7.839 4.55e-15 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 93.59 on 11 Df
Pseudo R-squared: 0.3191
Number of iterations: 19 (BFGS) + 1 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + ethnicity_cat2 +
religion_cat3 + target, data = d_sim_us_children)
Residuals:
Min 1Q Median 3Q Max
-1.2274 -0.5835 -0.1807 0.4714 2.1385
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.02034 0.11764 0.173 0.863267
scale(age) -0.21334 0.09342 -2.284 0.025836 *
gender_m 0.07626 0.10654 0.716 0.476799
ethnicity_cat2_POC -0.17057 0.11696 -1.458 0.149793
religion_cat3_christian 0.01444 0.14471 0.100 0.920839
religion_cat3_other -0.06421 0.18296 -0.351 0.726826
target1 -0.27972 0.30390 -0.920 0.360925
target2 0.07194 0.27838 0.258 0.796924
target3 0.29801 0.25626 1.163 0.249328
target4 0.13588 0.32133 0.423 0.673854
target5 -0.37378 0.29628 -1.262 0.211826
target6 -0.55706 0.30929 -1.801 0.076550 .
target7 -0.80145 0.31683 -2.530 0.013974 *
target8 -0.03422 0.29357 -0.117 0.907578
target9 1.00060 0.26667 3.752 0.000388 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8181 on 62 degrees of freedom
(40 observations deleted due to missingness)
Multiple R-squared: 0.4005, Adjusted R-squared: 0.2652
F-statistic: 2.959 on 14 and 62 DF, p-value: 0.001681Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + ethnicity_cat2 + religion_cat3 + +(1 | target)
Data: d_sim_us_children
REML criterion at convergence: 207.3
Scaled residuals:
Min 1Q Median 3Q Max
-1.4061 -0.7133 -0.2225 0.5368 2.3615
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.1930 0.4393
Residual 0.6683 0.8175
Number of obs: 77, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 0.01347 0.18137 11.62032 0.074 0.9421
scale(age) -0.22784 0.09187 65.64777 -2.480 0.0157 *
gender_m 0.03891 0.10262 69.18043 0.379 0.7057
ethnicity_cat2_POC -0.17952 0.11230 69.48885 -1.599 0.1145
religion_cat3_christian 0.01621 0.13990 68.79762 0.116 0.9081
religion_cat3_other -0.12055 0.17749 68.24187 -0.679 0.4993
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Correlation of Fixed Effects:
(Intr) scl(g) gndr_m e_2_PO rlgn_ct3_c
scale(age) -0.053
gender_m 0.048 -0.089
ethnc_2_POC -0.280 0.026 0.163
rlgn_ct3_ch -0.215 0.032 -0.123 0.071
rlgn_ct3_th 0.264 0.000 0.024 -0.118 -0.609
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + ethnicity_cat2 + religion_cat3 +
target, data = d_sim_us_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.8497 -0.8327 -0.1505 0.7703 2.7613
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.85837 0.07026 -12.218 < 2e-16 ***
scale(age) -0.15075 0.05500 -2.741 0.00612 **
gender_m 0.04090 0.06394 0.640 0.52241
ethnicity_cat2_POC -0.11247 0.06814 -1.651 0.09882 .
religion_cat3_christian 0.01392 0.08658 0.161 0.87230
religion_cat3_other -0.03952 0.11112 -0.356 0.72211
target1 -0.18171 0.18415 -0.987 0.32376
target2 0.08593 0.16244 0.529 0.59682
target3 0.18538 0.14817 1.251 0.21088
target4 0.11564 0.18959 0.610 0.54189
target5 -0.22607 0.18211 -1.241 0.21446
target6 -0.36906 0.18950 -1.948 0.05147 .
target7 -0.63194 0.21537 -2.934 0.00334 **
target8 0.01587 0.17338 0.092 0.92706
target9 0.63907 0.14801 4.318 1.58e-05 ***
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 19.646 3.107 6.322 2.57e-10 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 71.22 on 16 Df
Pseudo R-squared: 0.4194
Number of iterations: 25 (BFGS) + 2 (Fisher scoring) Ghana
Ghana adults
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_gh_adults)
Residuals:
Min 1Q Median 3Q Max
-1.2386 -0.8853 -0.3385 0.9651 1.8986
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.02698 0.08100 -0.333 0.740
scale(age) -0.19804 0.08128 -2.437 0.016 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9787 on 144 degrees of freedom
(4 observations deleted due to missingness)
Multiple R-squared: 0.0396, Adjusted R-squared: 0.03293
F-statistic: 5.937 on 1 and 144 DF, p-value: 0.01605
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_gh_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.3769 -0.9227 0.0304 0.9398 1.3404
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.36814 0.08410 -16.27 <2e-16 ***
scale(age) -0.15254 0.07625 -2.00 0.0455 *
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 4.8692 0.5593 8.705 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 91.73 on 3 Df
Pseudo R-squared: 0.02866
Number of iterations: 9 (BFGS) + 2 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_gh_adults)
Residuals:
Min 1Q Median 3Q Max
-1.0996 -0.8040 -0.2985 1.0829 1.7730
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.01774 0.08162 -0.217 0.8283
gender_m -0.14779 0.08162 -1.811 0.0722 .
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9924 on 148 degrees of freedom
Multiple R-squared: 0.02167, Adjusted R-squared: 0.01506
F-statistic: 3.279 on 1 and 148 DF, p-value: 0.07222
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_gh_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.1690 -0.8288 0.0811 0.9828 1.3189
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.35196 0.08386 -16.122 <2e-16 ***
gender_m -0.10810 0.07501 -1.441 0.15
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 4.7577 0.5371 8.858 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 90.6 on 3 Df
Pseudo R-squared: 0.01649
Number of iterations: 13 (BFGS) + 2 (Fisher scoring) Race/ethnicity
Call:
lm(formula = scale(MSE) ~ ethnicity_cat2, data = d_sim_gh_adults)
Residuals:
Min 1Q Median 3Q Max
-1.0429 -0.8565 -0.3436 1.0072 1.7045
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.01987 0.08351 -0.238 0.812
ethnicity_cat2_nonFante -0.09316 0.08351 -1.116 0.266
Residual standard error: 0.9992 on 148 degrees of freedom
Multiple R-squared: 0.00834, Adjusted R-squared: 0.00164
F-statistic: 1.245 on 1 and 148 DF, p-value: 0.2664
Call:
betareg(formula = MSE_rescaled ~ ethnicity_cat2, data = d_sim_gh_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.0985 -0.8757 0.0208 0.9482 1.2950
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.35270 0.08523 -15.871 <2e-16 ***
ethnicity_cat2_nonFante -0.07164 0.07638 -0.938 0.348
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 4.7180 0.5324 8.862 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 90.01 on 3 Df
Pseudo R-squared: 0.00706
Number of iterations: 12 (BFGS) + 1 (Fisher scoring) Education
Call:
lm(formula = scale(MSE) ~ scale(education_catX), data = d_sim_gh_adults %>%
mutate(education_catX = as.numeric(education_catX)))
Residuals:
Min 1Q Median 3Q Max
-1.0573 -0.9852 -0.2745 1.0859 1.8021
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.006507 0.081823 0.08 0.937
scale(education_catX) -0.097731 0.082099 -1.19 0.236
Residual standard error: 0.9988 on 147 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.009548, Adjusted R-squared: 0.00281
F-statistic: 1.417 on 1 and 147 DF, p-value: 0.2358
Call:
betareg(formula = MSE_rescaled ~ scale(education_catX), data = d_sim_gh_adults %>%
mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.1354 -1.0310 0.0899 1.0110 1.3492
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.33179 0.08327 -15.99 <2e-16 ***
scale(education_catX) -0.08611 0.07556 -1.14 0.254
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 4.7304 0.5353 8.838 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 88.98 on 3 Df
Pseudo R-squared: 0.01024
Number of iterations: 12 (BFGS) + 1 (Fisher scoring) 
Call:
lm(formula = scale(MSE) ~ education_cat2, data = d_sim_gh_adults)
Residuals:
Min 1Q Median 3Q Max
-1.1159 -0.8587 -0.3149 1.0742 1.8406
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.01773 0.08185 0.217 0.829
education_cat2_hs -0.12862 0.08185 -1.572 0.118
Residual standard error: 0.9953 on 147 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.01652, Adjusted R-squared: 0.009833
F-statistic: 2.47 on 1 and 147 DF, p-value: 0.1182
Call:
betareg(formula = MSE_rescaled ~ education_cat2, data = d_sim_gh_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.2261 -0.8659 0.0547 0.9816 1.3662
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.32323 0.08306 -15.931 <2e-16 ***
education_cat2_hs -0.11238 0.07457 -1.507 0.132
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 4.7623 0.5391 8.834 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 89.46 on 3 Df
Pseudo R-squared: 0.01759
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) Rural/urban
Call:
lm(formula = scale(MSE) ~ urban_rural_cat2, data = d_sim_gh_adults)
Residuals:
Min 1Q Median 3Q Max
-1.1431 -0.8662 -0.3567 1.0469 1.8466
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.03508 0.08392 0.418 0.677
urban_rural_cat2_rural 0.13849 0.08392 1.650 0.101
Residual standard error: 0.9943 on 148 degrees of freedom
Multiple R-squared: 0.01807, Adjusted R-squared: 0.01143
F-statistic: 2.723 on 1 and 148 DF, p-value: 0.101
Call:
betareg(formula = MSE_rescaled ~ urban_rural_cat2, data = d_sim_gh_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.2964 -0.8646 0.0106 0.9582 1.3764
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.30662 0.08407 -15.542 <2e-16 ***
urban_rural_cat2_rural 0.13297 0.07610 1.747 0.0806 .
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 4.7883 0.5408 8.855 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 91.05 on 3 Df
Pseudo R-squared: 0.02284
Number of iterations: 11 (BFGS) + 2 (Fisher scoring) Religion
Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_gh_adults, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-1.1893 -0.2391 0.0000 0.1387 1.9963
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -4.532e-18 4.553e-02 0.000 1.00000
target1 -9.696e-01 1.366e-01 -7.099 5.78e-11 ***
target2 -9.696e-01 1.366e-01 -7.099 5.78e-11 ***
target3 -2.562e-01 1.366e-01 -1.876 0.06273 .
target4 1.855e-01 1.366e-01 1.358 0.17661
target5 5.517e-01 1.366e-01 4.039 8.80e-05 ***
target6 1.246e+00 1.366e-01 9.125 7.07e-16 ***
target7 9.162e-02 1.366e-01 0.671 0.50344
target8 -9.374e-01 1.366e-01 -6.863 2.00e-10 ***
target9 -4.058e-01 1.366e-01 -2.971 0.00349 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.5576 on 140 degrees of freedom
Multiple R-squared: 0.7079, Adjusted R-squared: 0.6891
F-statistic: 37.69 on 9 and 140 DF, p-value: < 2.2e-16
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_gh_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.3635 -0.3226 0.0000 0.3134 2.7090
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.56482 0.05624 -27.822 < 2e-16 ***
target1 -1.13061 0.19385 -5.833 5.46e-09 ***
target2 -1.13061 0.19385 -5.833 5.46e-09 ***
target3 -0.14179 0.15280 -0.928 0.353433
target4 0.41155 0.13631 3.019 0.002535 **
target5 0.76984 0.12911 5.963 2.48e-09 ***
target6 1.29162 0.12366 10.445 < 2e-16 ***
target7 0.12314 0.14412 0.854 0.392858
target8 -1.07679 0.19147 -5.624 1.87e-08 ***
target9 -0.56177 0.16906 -3.323 0.000891 ***
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 16.73 1.95 8.58 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 178.4 on 11 Df
Pseudo R-squared: 0.7305
Number of iterations: 21 (BFGS) + 2 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + scale(education_catX) +
ethnicity_cat2 + urban_rural_cat2 + target, data = d_sim_gh_adults %>%
mutate(education_catX = as.numeric(education_catX)))
Residuals:
Min 1Q Median 3Q Max
-1.12346 -0.31047 -0.02467 0.17093 1.93057
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.003505 0.049807 -0.070 0.944005
scale(age) -0.086753 0.052147 -1.664 0.098598 .
gender_m 0.017111 0.049914 0.343 0.732291
scale(education_catX) -0.004249 0.055438 -0.077 0.939026
ethnicity_cat2_nonFante -0.069933 0.052244 -1.339 0.183044
urban_rural_cat2_rural 0.063068 0.060166 1.048 0.296471
target1 -0.873415 0.144950 -6.026 1.62e-08 ***
target2 -0.977133 0.140261 -6.967 1.46e-10 ***
target3 -0.243990 0.142003 -1.718 0.088139 .
target4 0.149587 0.142735 1.048 0.296582
target5 0.553041 0.146034 3.787 0.000232 ***
target6 1.163982 0.148872 7.819 1.60e-12 ***
target7 0.132432 0.141017 0.939 0.349408
target8 -0.936736 0.144435 -6.486 1.69e-09 ***
target9 -0.409953 0.142470 -2.877 0.004688 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.5616 on 130 degrees of freedom
(5 observations deleted due to missingness)
Multiple R-squared: 0.7127, Adjusted R-squared: 0.6818
F-statistic: 23.04 on 14 and 130 DF, p-value: < 2.2e-16
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + education_cat2 +
ethnicity_cat2 + urban_rural_cat2 + target, data = d_sim_gh_adults)
Residuals:
Min 1Q Median 3Q Max
-1.12477 -0.31513 -0.02035 0.16258 1.92649
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.003320 0.049643 -0.067 0.946776
scale(age) -0.088090 0.052181 -1.688 0.093779 .
gender_m 0.016507 0.050079 0.330 0.742216
education_cat2_hs -0.009189 0.058030 -0.158 0.874432
ethnicity_cat2_nonFante -0.069994 0.051216 -1.367 0.174097
urban_rural_cat2_rural 0.060454 0.060542 0.999 0.319869
target1 -0.873195 0.144688 -6.035 1.55e-08 ***
target2 -0.976031 0.140221 -6.961 1.50e-10 ***
target3 -0.245123 0.142185 -1.724 0.087090 .
target4 0.148502 0.142938 1.039 0.300768
target5 0.553701 0.145771 3.798 0.000223 ***
target6 1.160870 0.150178 7.730 2.58e-12 ***
target7 0.136151 0.143523 0.949 0.344569
target8 -0.936657 0.144390 -6.487 1.67e-09 ***
target9 -0.409053 0.141656 -2.888 0.004547 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.5616 on 130 degrees of freedom
(5 observations deleted due to missingness)
Multiple R-squared: 0.7127, Adjusted R-squared: 0.6818
F-statistic: 23.04 on 14 and 130 DF, p-value: < 2.2e-16Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + scale(education_catX) + ethnicity_cat2 +
urban_rural_cat2 + +(1 | target)
Data: d_sim_gh_adults %>% mutate(education_catX = as.numeric(education_catX))
REML criterion at convergence: 294.6
Scaled residuals:
Min 1Q Median 3Q Max
-2.0041 -0.5266 -0.0655 0.3215 3.3997
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.7094 0.8423
Residual 0.3154 0.5616
Number of obs: 145, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) -0.003899 0.270967 9.030693 -0.014 0.9888
scale(age) -0.091131 0.052056 130.892796 -1.751 0.0824 .
gender_m 0.013662 0.049860 130.549886 0.274 0.7845
scale(education_catX) -0.003250 0.055403 130.321271 -0.059 0.9533
ethnicity_cat2_nonFante -0.072511 0.052151 130.907198 -1.390 0.1668
urban_rural_cat2_rural 0.067259 0.060085 130.686246 1.119 0.2650
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Correlation of Fixed Effects:
(Intr) scl(g) gndr_m sc(_X) et_2_F
scale(age) -0.011
gender_m 0.021 -0.220
scl(dctn_X) 0.015 -0.104 0.015
ethncty_2_F 0.029 0.106 -0.047 -0.221
urbn_rrl_2_ 0.046 -0.250 0.074 0.508 -0.219Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + education_cat2 + ethnicity_cat2 +
urban_rural_cat2 + +(1 | target)
Data: d_sim_gh_adults
REML criterion at convergence: 294.5
Scaled residuals:
Min 1Q Median 3Q Max
-2.0063 -0.5342 -0.0673 0.3121 3.3914
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.7079 0.8413
Residual 0.3154 0.5616
Number of obs: 145, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) -0.003822 0.270647 9.023536 -0.014 0.9890
scale(age) -0.092627 0.052084 130.941922 -1.778 0.0777 .
gender_m 0.012855 0.050023 130.574497 0.257 0.7976
education_cat2_hs -0.011810 0.057958 130.636468 -0.204 0.8388
ethnicity_cat2_nonFante -0.072131 0.051127 130.889768 -1.411 0.1607
urban_rural_cat2_rural 0.062685 0.060469 130.615654 1.037 0.3018
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Correlation of Fixed Effects:
(Intr) scl(g) gndr_m edc_2_ et_2_F
scale(age) -0.009
gender_m 0.020 -0.208
edctn_ct2_h 0.003 0.110 0.083
ethncty_2_F 0.033 0.073 -0.053 -0.102
urbn_rrl_2_ 0.039 -0.139 0.108 0.517 -0.161
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + scale(education_catX) + ethnicity_cat2 +
urban_rural_cat2 + target, data = d_sim_gh_adults %>% mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.9169 -0.3454 -0.0051 0.3780 2.7670
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.570784 0.059515 -26.393 < 2e-16 ***
scale(age) -0.076474 0.053960 -1.417 0.15641
gender_m 0.033911 0.053190 0.638 0.52376
scale(education_catX) 0.001766 0.059957 0.029 0.97650
ethnicity_cat2_nonFante -0.076939 0.056003 -1.374 0.16950
urban_rural_cat2_rural 0.053495 0.063147 0.847 0.39691
target1 -1.048075 0.198822 -5.271 1.35e-07 ***
target2 -1.155938 0.195639 -5.909 3.45e-09 ***
target3 -0.105359 0.155715 -0.677 0.49865
target4 0.378160 0.141766 2.667 0.00764 **
target5 0.769768 0.136250 5.650 1.61e-08 ***
target6 1.225444 0.132726 9.233 < 2e-16 ***
target7 0.157647 0.146766 1.074 0.28276
target8 -1.094933 0.199131 -5.499 3.83e-08 ***
target9 -0.561446 0.171753 -3.269 0.00108 **
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 16.991 2.016 8.43 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 174.8 on 16 Df
Pseudo R-squared: 0.7326
Number of iterations: 26 (BFGS) + 2 (Fisher scoring)
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + education_cat2 + ethnicity_cat2 +
urban_rural_cat2 + target, data = d_sim_gh_adults %>% mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.9198 -0.3344 -0.0054 0.3677 2.7679
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.570914 0.059406 -26.444 < 2e-16 ***
scale(age) -0.076646 0.053948 -1.421 0.15540
gender_m 0.033760 0.053372 0.633 0.52704
education_cat2_hs -0.003045 0.060261 -0.051 0.95970
ethnicity_cat2_nonFante -0.076097 0.054146 -1.405 0.15990
urban_rural_cat2_rural 0.050902 0.062320 0.817 0.41405
target1 -1.048406 0.198568 -5.280 1.29e-07 ***
target2 -1.154696 0.195507 -5.906 3.50e-09 ***
target3 -0.105764 0.155956 -0.678 0.49767
target4 0.377334 0.142015 2.657 0.00788 **
target5 0.770512 0.135931 5.668 1.44e-08 ***
target6 1.224527 0.134262 9.120 < 2e-16 ***
target7 0.159591 0.149404 1.068 0.28544
target8 -1.095167 0.199094 -5.501 3.78e-08 ***
target9 -0.562289 0.171083 -3.287 0.00101 **
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 16.991 2.016 8.43 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 174.8 on 16 Df
Pseudo R-squared: 0.7326
Number of iterations: 25 (BFGS) + 2 (Fisher scoring) Ghana children
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_gh_children)
Residuals:
Min 1Q Median 3Q Max
-1.4019 -0.9130 0.0356 0.9492 1.5357
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.991e-18 8.187e-02 0.000 1.000
scale(age) 3.661e-02 8.214e-02 0.446 0.656
Residual standard error: 1.003 on 148 degrees of freedom
Multiple R-squared: 0.00134, Adjusted R-squared: -0.005407
F-statistic: 0.1986 on 1 and 148 DF, p-value: 0.6565
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_gh_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.6645 -0.7918 0.2020 0.9170 1.3511
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.84640 0.06194 -13.664 <2e-16 ***
scale(age) 0.02646 0.06047 0.438 0.662
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 7.181 0.789 9.102 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 70.61 on 3 Df
Pseudo R-squared: 0.001423
Number of iterations: 12 (BFGS) + 1 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_gh_children)
Residuals:
Min 1Q Median 3Q Max
-1.39196 -0.92322 0.04008 0.92426 1.58300
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.0006675 0.0820354 0.008 0.994
gender_m 0.0125151 0.0820354 0.153 0.879
Residual standard error: 1.003 on 148 degrees of freedom
Multiple R-squared: 0.0001572, Adjusted R-squared: -0.006598
F-statistic: 0.02327 on 1 and 148 DF, p-value: 0.879
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_gh_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.7201 -0.7971 0.2201 0.8948 1.3743
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.846241 0.062057 -13.637 <2e-16 ***
gender_m 0.000961 0.060344 0.016 0.987
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 7.1715 0.7879 9.103 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 70.51 on 3 Df
Pseudo R-squared: 1.744e-06
Number of iterations: 13 (BFGS) + 1 (Fisher scoring) Religion
Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_gh_children, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-1.83250 -0.38281 -0.01779 0.40059 1.88660
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.01057 0.05575 -0.190 0.84992
target1 -1.14561 0.16712 -6.855 2.09e-10 ***
target2 -0.38840 0.16712 -2.324 0.02156 *
target3 0.53385 0.16241 3.287 0.00128 **
target4 0.48096 0.16241 2.961 0.00360 **
target5 0.33720 0.17234 1.957 0.05238 .
target6 0.29145 0.16712 1.744 0.08335 .
target7 -0.90761 0.17234 -5.266 5.13e-07 ***
target8 -0.92673 0.16712 -5.545 1.41e-07 ***
target9 0.49460 0.16712 2.960 0.00362 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.6822 on 140 degrees of freedom
Multiple R-squared: 0.5627, Adjusted R-squared: 0.5346
F-statistic: 20.02 on 9 and 140 DF, p-value: < 2.2e-16
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_gh_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.5167 -0.4419 0.0127 0.6824 2.5324
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.91960 0.04531 -20.296 < 2e-16 ***
target1 -0.98240 0.15907 -6.176 6.58e-10 ***
target2 -0.33924 0.13679 -2.480 0.013136 *
target3 0.44120 0.11841 3.726 0.000194 ***
target4 0.42064 0.11862 3.546 0.000391 ***
target5 0.31662 0.12698 2.493 0.012649 *
target6 0.30578 0.12336 2.479 0.013183 *
target7 -0.70853 0.15334 -4.621 3.83e-06 ***
target8 -0.73387 0.14951 -4.908 9.18e-07 ***
target9 0.35644 0.12271 2.905 0.003675 **
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 16.447 1.863 8.83 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 131.6 on 11 Df
Pseudo R-squared: 0.5768
Number of iterations: 19 (BFGS) + 2 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + target, data = d_sim_gh_children)
Residuals:
Min 1Q Median 3Q Max
-1.99789 -0.33258 -0.04549 0.39728 1.74140
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.01525 0.05570 -0.274 0.78460
scale(age) -0.03187 0.05793 -0.550 0.58308
gender_m -0.09747 0.06051 -1.611 0.10948
target1 -1.17793 0.16805 -7.009 9.70e-11 ***
target2 -0.37780 0.16688 -2.264 0.02514 *
target3 0.50802 0.16330 3.111 0.00227 **
target4 0.44961 0.16323 2.754 0.00667 **
target5 0.34004 0.17200 1.977 0.05004 .
target6 0.30205 0.16688 1.810 0.07249 .
target7 -0.89085 0.17232 -5.170 8.06e-07 ***
target8 -0.97009 0.16890 -5.744 5.67e-08 ***
target9 0.55008 0.17036 3.229 0.00155 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.6807 on 138 degrees of freedom
Multiple R-squared: 0.5709, Adjusted R-squared: 0.5367
F-statistic: 16.69 on 11 and 138 DF, p-value: < 2.2e-16Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + +(1 | target)
Data: d_sim_gh_children
REML criterion at convergence: 346.3
Scaled residuals:
Min 1Q Median 3Q Max
-2.87483 -0.53742 -0.06315 0.61199 2.54883
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.6162 0.7850
Residual 0.4634 0.6807
Number of obs: 150, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) -0.01447 0.25441 8.97306 -0.057 0.956
scale(age) -0.02814 0.05790 138.26565 -0.486 0.628
gender_m -0.09105 0.06034 139.46246 -1.509 0.134
Correlation of Fixed Effects:
(Intr) scl(g)
scale(age) 0.003
gender_m 0.011 0.252
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + target, data = d_sim_gh_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.8542 -0.4865 -0.0147 0.7125 2.4770
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.92257 0.04514 -20.439 < 2e-16 ***
scale(age) -0.01542 0.04480 -0.344 0.730788
gender_m -0.06776 0.04665 -1.453 0.146329
target1 -1.00246 0.15895 -6.307 2.85e-10 ***
target2 -0.32517 0.13587 -2.393 0.016698 *
target3 0.42091 0.11871 3.546 0.000391 ***
target4 0.39827 0.11885 3.351 0.000805 ***
target5 0.31951 0.12626 2.531 0.011385 *
target6 0.31074 0.12278 2.531 0.011376 *
target7 -0.69862 0.15272 -4.574 4.78e-06 ***
target8 -0.76168 0.15010 -5.074 3.89e-07 ***
target9 0.38987 0.12499 3.119 0.001813 **
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 16.670 1.888 8.828 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 132.6 on 13 Df
Pseudo R-squared: 0.586
Number of iterations: 20 (BFGS) + 2 (Fisher scoring) Thailand
Thailand adults
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_th_adults)
Residuals:
Min 1Q Median 3Q Max
-1.8227 -0.7161 -0.1706 0.4121 2.3409
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.005132 0.080033 -0.064 0.9490
scale(age) 0.234203 0.080302 2.917 0.0041 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9769 on 147 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.0547, Adjusted R-squared: 0.04827
F-statistic: 8.506 on 1 and 147 DF, p-value: 0.004095
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_th_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.3514 -0.6759 -0.0719 0.4979 2.1050
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.80410 0.05074 -15.846 < 2e-16 ***
scale(age) 0.13562 0.04945 2.743 0.00609 **
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 11.192 1.252 8.939 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 94.7 on 3 Df
Pseudo R-squared: 0.04855
Number of iterations: 13 (BFGS) + 1 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_th_adults)
Residuals:
Min 1Q Median 3Q Max
-1.5220 -0.7040 -0.2978 0.4676 2.2403
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.3763 0.2408 -1.563 0.1203
gender_m 0.4819 0.2520 1.912 0.0578 .
gender_o -0.8133 0.4720 -1.723 0.0870 .
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9943 on 147 degrees of freedom
Multiple R-squared: 0.02468, Adjusted R-squared: 0.01141
F-statistic: 1.86 on 2 and 147 DF, p-value: 0.1594contrasts dropped from factor gender due to missing levels
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_th_adults %>%
filter(gender != "other"), contrasts = list(gender = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-1.5274 -0.7182 -0.3045 0.4955 2.2484
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.01429 0.08376 0.171 0.865
gender1 -0.07553 0.08376 -0.902 0.369
Residual standard error: 1.001 on 146 degrees of freedom
Multiple R-squared: 0.005538, Adjusted R-squared: -0.001273
F-statistic: 0.8131 on 1 and 146 DF, p-value: 0.3687
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_th_adults %>% filter(gender !=
"other") %>% mutate(gender = case_when(gender == "female" ~ -1, gender == "male" ~
1)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.8734 -0.6511 -0.1933 0.5496 1.9995
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.78071 0.05220 -14.956 <2e-16 ***
gender 0.03438 0.05143 0.668 0.504
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 10.843 1.215 8.923 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 91.08 on 3 Df
Pseudo R-squared: 0.003166
Number of iterations: 11 (BFGS) + 2 (Fisher scoring) Race/ethnicity
Education
Call:
lm(formula = scale(MSE) ~ scale(education_catX), data = d_sim_th_adults %>%
mutate(education_catX = as.numeric(education_catX)))
Residuals:
Min 1Q Median 3Q Max
-1.6792 -0.6830 -0.2081 0.4750 2.3728
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.03157 0.08104 -0.390 0.6975
scale(education_catX) -0.18402 0.08132 -2.263 0.0252 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9759 on 143 degrees of freedom
(5 observations deleted due to missingness)
Multiple R-squared: 0.03457, Adjusted R-squared: 0.02782
F-statistic: 5.12 on 1 and 143 DF, p-value: 0.02515
Call:
betareg(formula = MSE_rescaled ~ scale(education_catX), data = d_sim_th_adults %>%
mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.1629 -0.6706 -0.1095 0.5518 2.1262
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.82002 0.05155 -15.907 <2e-16 ***
scale(education_catX) -0.10403 0.05061 -2.055 0.0398 *
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 11.20 1.27 8.817 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 92.59 on 3 Df
Pseudo R-squared: 0.02906
Number of iterations: 13 (BFGS) + 2 (Fisher scoring) 
Call:
lm(formula = scale(MSE) ~ education_cat2, data = d_sim_th_adults)
Residuals:
Min 1Q Median 3Q Max
-1.6413 -0.7070 -0.2242 0.4714 2.3545
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.001235 0.082432 -0.015 0.9881
education_cat2_coll -0.175928 0.082432 -2.134 0.0345 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9777 on 143 degrees of freedom
(5 observations deleted due to missingness)
Multiple R-squared: 0.03087, Adjusted R-squared: 0.02409
F-statistic: 4.555 on 1 and 143 DF, p-value: 0.03453
Call:
betareg(formula = MSE_rescaled ~ education_cat2, data = d_sim_th_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.1006 -0.6261 -0.0953 0.5436 2.1057
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.80310 0.05212 -15.408 <2e-16 ***
education_cat2_coll -0.09583 0.05132 -1.867 0.0619 .
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 11.142 1.263 8.818 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 92.26 on 3 Df
Pseudo R-squared: 0.0246
Number of iterations: 10 (BFGS) + 2 (Fisher scoring) Rural/urban
Call:
lm(formula = scale(MSE) ~ urban_rural_cat2, data = d_sim_th_adults)
Residuals:
Min 1Q Median 3Q Max
-1.4677 -0.7626 -0.2797 0.4761 2.3065
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.03534 0.08900 -0.397 0.692
urban_rural_cat2_rural 0.07578 0.08900 0.851 0.396
Residual standard error: 0.991 on 139 degrees of freedom
(9 observations deleted due to missingness)
Multiple R-squared: 0.005188, Adjusted R-squared: -0.001969
F-statistic: 0.7249 on 1 and 139 DF, p-value: 0.396
Call:
betareg(formula = MSE_rescaled ~ urban_rural_cat2, data = d_sim_th_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.8613 -0.7207 -0.1704 0.5443 2.0901
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.82040 0.05611 -14.622 <2e-16 ***
urban_rural_cat2_rural 0.04982 0.05529 0.901 0.367
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 11.031 1.268 8.702 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 88.38 on 3 Df
Pseudo R-squared: 0.006162
Number of iterations: 8 (BFGS) + 2 (Fisher scoring) Religion
Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_th_adults, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-1.5241 -0.7322 -0.1309 0.6421 2.3396
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -3.082e-16 8.053e-02 0.000 1.00000
target1 2.499e-01 2.416e-01 1.034 0.30281
target2 2.884e-02 2.416e-01 0.119 0.90515
target3 1.968e-01 2.416e-01 0.815 0.41672
target4 -1.156e-01 2.416e-01 -0.478 0.63311
target5 3.514e-01 2.416e-01 1.454 0.14806
target6 -1.622e-01 2.416e-01 -0.672 0.50300
target7 -6.329e-01 2.416e-01 -2.620 0.00977 **
target8 1.830e-01 2.416e-01 0.758 0.44997
target9 2.243e-01 2.416e-01 0.929 0.35474
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9863 on 140 degrees of freedom
Multiple R-squared: 0.08599, Adjusted R-squared: 0.02723
F-statistic: 1.464 on 9 and 140 DF, p-value: 0.1672
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_th_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.8707 -0.7335 -0.0492 0.7037 2.2438
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.803257 0.049709 -16.159 <2e-16 ***
target1 0.178585 0.143284 1.246 0.2126
target2 -0.008229 0.146551 -0.056 0.9552
target3 0.111397 0.144360 0.772 0.4403
target4 -0.127666 0.149085 -0.856 0.3918
target5 0.216832 0.142722 1.519 0.1287
target6 -0.119940 0.148911 -0.805 0.4206
target7 -0.380804 0.155562 -2.448 0.0144 *
target8 0.146666 0.143782 1.020 0.3077
target9 0.132666 0.144008 0.921 0.3569
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 11.663 1.302 8.956 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 98.26 on 11 Df
Pseudo R-squared: 0.08959
Number of iterations: 20 (BFGS) + 2 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + scale(education_catX) +
urban_rural_cat2 + target, data = d_sim_th_adults %>% filter(gender !=
"other") %>% mutate(education_catX = as.numeric(education_catX),
gender = case_when(gender == "female" ~ -1, gender == "male" ~
1)))
Residuals:
Min 1Q Median 3Q Max
-1.67271 -0.63947 -0.06236 0.46574 2.46055
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.039643 0.089408 -0.443 0.6583
scale(age) 0.188059 0.100709 1.867 0.0643 .
gender 0.038295 0.090801 0.422 0.6740
scale(education_catX) -0.106247 0.102039 -1.041 0.2999
urban_rural_cat2_rural -0.003657 0.092223 -0.040 0.9684
target1 0.272652 0.249196 1.094 0.2761
target2 -0.129426 0.264834 -0.489 0.6259
target3 0.345354 0.239768 1.440 0.1524
target4 -0.009965 0.246891 -0.040 0.9679
target5 0.283155 0.254478 1.113 0.2681
target6 -0.072453 0.246277 -0.294 0.7691
target7 -0.499755 0.244246 -2.046 0.0429 *
target8 0.224240 0.246431 0.910 0.3647
target9 -0.114037 0.287353 -0.397 0.6922
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9548 on 120 degrees of freedom
(14 observations deleted due to missingness)
Multiple R-squared: 0.1403, Adjusted R-squared: 0.04712
F-statistic: 1.506 on 13 and 120 DF, p-value: 0.1247
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + education_cat2 +
urban_rural_cat2 + target, data = d_sim_th_adults %>% filter(gender !=
"other") %>% mutate(education_catX = as.numeric(education_catX),
gender = case_when(gender == "female" ~ -1, gender == "male" ~
1)))
Residuals:
Min 1Q Median 3Q Max
-1.64313 -0.61016 -0.03672 0.45523 2.43583
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.024350 0.093046 -0.262 0.7940
scale(age) 0.199076 0.105015 1.896 0.0604 .
gender 0.034180 0.090981 0.376 0.7078
education_cat2_coll -0.078892 0.109726 -0.719 0.4735
urban_rural_cat2_rural -0.005729 0.094139 -0.061 0.9516
target1 0.270855 0.251209 1.078 0.2831
target2 -0.128616 0.269188 -0.478 0.6337
target3 0.338462 0.240205 1.409 0.1614
target4 -0.026638 0.246474 -0.108 0.9141
target5 0.289487 0.254977 1.135 0.2585
target6 -0.069188 0.246847 -0.280 0.7797
target7 -0.497339 0.245189 -2.028 0.0447 *
target8 0.218400 0.246959 0.884 0.3783
target9 -0.101148 0.287542 -0.352 0.7256
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.957 on 120 degrees of freedom
(14 observations deleted due to missingness)
Multiple R-squared: 0.1362, Adjusted R-squared: 0.04264
F-statistic: 1.456 on 13 and 120 DF, p-value: 0.1443Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + scale(education_catX) + urban_rural_cat2 +
+(1 | target)
Data:
d_sim_th_adults %>% filter(gender != "other") %>% mutate(education_catX = as.numeric(education_catX),
gender = case_when(gender == "female" ~ -1, gender == "male" ~ 1))
REML criterion at convergence: 379.4
Scaled residuals:
Min 1Q Median 3Q Max
-1.6687 -0.7160 -0.1924 0.3970 2.5214
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.01596 0.1263
Residual 0.90806 0.9529
Number of obs: 134, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) -0.038622 0.097341 11.806262 -0.397 0.6986
scale(age) 0.195337 0.098909 126.599653 1.975 0.0505 .
gender 0.027442 0.085586 125.968517 0.321 0.7490
scale(education_catX) -0.083493 0.098973 128.583490 -0.844 0.4005
urban_rural_cat2_rural 0.002382 0.090489 126.740314 0.026 0.9790
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Correlation of Fixed Effects:
(Intr) scl(g) gender sc(_X)
scale(age) 0.017
gender 0.122 -0.098
scl(dctn_X) -0.036 0.507 -0.028
urbn_rrl_2_ -0.302 -0.078 0.102 0.133Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + education_cat2 + urban_rural_cat2 +
+(1 | target)
Data:
d_sim_th_adults %>% filter(gender != "other") %>% mutate(education_catX = as.numeric(education_catX),
gender = case_when(gender == "female" ~ -1, gender == "male" ~ 1))
REML criterion at convergence: 379.7
Scaled residuals:
Min 1Q Median 3Q Max
-1.7145 -0.7209 -0.1902 0.4080 2.4959
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.01459 0.1208
Residual 0.91206 0.9550
Number of obs: 134, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) -0.028199 0.099866 13.347738 -0.282 0.7820
scale(age) 0.206975 0.102633 127.561797 2.017 0.0458 *
gender 0.024397 0.085675 125.726454 0.285 0.7763
education_cat2_coll -0.056640 0.105293 128.975954 -0.538 0.5915
urban_rural_cat2_rural 0.002019 0.091958 127.607802 0.022 0.9825
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Correlation of Fixed Effects:
(Intr) scl(g) gender edc_2_
scale(age) -0.105
gender 0.114 -0.071
edctn_ct2_c -0.248 0.555 0.019
urbn_rrl_2_ -0.339 -0.020 0.109 0.212
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + scale(education_catX) + urban_rural_cat2 +
target, data = d_sim_th_adults %>% filter(gender != "other") %>% mutate(education_catX = as.numeric(education_catX),
gender = case_when(gender == "female" ~ -1, gender == "male" ~ 1)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.0637 -0.6569 -0.0149 0.5831 2.4739
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.821619 0.054327 -15.124 <2e-16 ***
scale(age) 0.114569 0.059649 1.921 0.0548 .
gender 0.015689 0.054344 0.289 0.7728
scale(education_catX) -0.061299 0.061070 -1.004 0.3155
urban_rural_cat2_rural -0.005483 0.055405 -0.099 0.9212
target1 0.176979 0.146268 1.210 0.2263
target2 -0.124730 0.160882 -0.775 0.4382
target3 0.204585 0.141089 1.450 0.1470
target4 -0.065861 0.148931 -0.442 0.6583
target5 0.189076 0.148649 1.272 0.2034
target6 -0.051661 0.148111 -0.349 0.7272
target7 -0.293506 0.153736 -1.909 0.0562 .
target8 0.164382 0.144564 1.137 0.2555
target9 -0.055437 0.171422 -0.323 0.7464
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 13.058 1.548 8.434 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 95.14 on 15 Df
Pseudo R-squared: 0.1334
Number of iterations: 23 (BFGS) + 2 (Fisher scoring)
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + education_cat2 + urban_rural_cat2 +
target, data = d_sim_th_adults %>% filter(gender != "other") %>% mutate(education_catX = as.numeric(education_catX),
gender = case_when(gender == "female" ~ -1, gender == "male" ~ 1)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.0464 -0.6345 0.0231 0.5641 2.4493
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.813684 0.056382 -14.432 <2e-16 ***
scale(age) 0.122737 0.062113 1.976 0.0482 *
gender 0.012987 0.054445 0.239 0.8115
education_cat2_coll -0.041505 0.065663 -0.632 0.5273
urban_rural_cat2_rural -0.005253 0.056586 -0.093 0.9260
target1 0.174669 0.147408 1.185 0.2360
target2 -0.123914 0.163377 -0.758 0.4482
target3 0.200468 0.141291 1.419 0.1559
target4 -0.077748 0.148624 -0.523 0.6009
target5 0.192875 0.148814 1.296 0.1949
target6 -0.047675 0.148346 -0.321 0.7479
target7 -0.291774 0.154209 -1.892 0.0585 .
target8 0.160173 0.144838 1.106 0.2688
target9 -0.047379 0.171469 -0.276 0.7823
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 12.997 1.541 8.435 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 94.83 on 15 Df
Pseudo R-squared: 0.1296
Number of iterations: 24 (BFGS) + 2 (Fisher scoring) Thailand children
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_th_children)
Residuals:
Min 1Q Median 3Q Max
-1.5399 -0.7272 -0.2931 0.7147 2.9228
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 4.658e-17 8.100e-02 0.000 1.000
scale(age) -9.654e-02 8.127e-02 -1.188 0.237
Residual standard error: 0.9986 on 150 degrees of freedom
Multiple R-squared: 0.00932, Adjusted R-squared: 0.002716
F-statistic: 1.411 on 1 and 150 DF, p-value: 0.2367
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_th_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.9606 -0.6910 -0.1969 0.7630 2.6049
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.93266 0.04237 -22.01 <2e-16 ***
scale(age) -0.05117 0.04194 -1.22 0.222
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 17.059 1.913 8.916 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 129.1 on 3 Df
Pseudo R-squared: 0.01085
Number of iterations: 9 (BFGS) + 2 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_th_children)
Residuals:
Min 1Q Median 3Q Max
-1.6021 -0.7579 -0.1796 0.7213 3.1169
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.008783 0.081551 0.108 0.914
gender_m 0.083441 0.081551 1.023 0.308
Residual standard error: 0.9998 on 150 degrees of freedom
Multiple R-squared: 0.006931, Adjusted R-squared: 0.0003104
F-statistic: 1.047 on 1 and 150 DF, p-value: 0.3079
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_th_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.0166 -0.7391 -0.0938 0.7622 2.7135
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.92876 0.04264 -21.782 <2e-16 ***
gender_m 0.03456 0.04205 0.822 0.411
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 16.964 1.902 8.918 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 128.7 on 3 Df
Pseudo R-squared: 0.004708
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) Religion
Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_th_children, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-1.5939 -0.6810 -0.2017 0.4187 2.8156
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.001869 0.078903 -0.024 0.9811
target1 -0.022011 0.238036 -0.092 0.9265
target2 -0.231547 0.231320 -1.001 0.3185
target3 -0.019541 0.238036 -0.082 0.9347
target4 -0.276511 0.238036 -1.162 0.2473
target5 -0.352161 0.238036 -1.479 0.1412
target6 -0.109907 0.238036 -0.462 0.6450
target7 -0.399233 0.238036 -1.677 0.0957 .
target8 0.452051 0.238036 1.899 0.0596 .
target9 0.515583 0.231320 2.229 0.0274 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9725 on 142 degrees of freedom
Multiple R-squared: 0.1107, Adjusted R-squared: 0.05433
F-statistic: 1.964 on 9 and 142 DF, p-value: 0.04774
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_th_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.8769 -0.7085 -0.1184 0.5032 2.7180
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.939353 0.040478 -23.206 <2e-16 ***
target1 0.001302 0.120112 0.011 0.9914
target2 -0.096821 0.118742 -0.815 0.4148
target3 -0.016916 0.120483 -0.140 0.8883
target4 -0.163343 0.123739 -1.320 0.1868
target5 -0.168808 0.123870 -1.363 0.1729
target6 -0.048193 0.121139 -0.398 0.6908
target7 -0.232307 0.125437 -1.852 0.0640 .
target8 0.231012 0.116053 1.991 0.0465 *
target9 0.237074 0.112724 2.103 0.0355 *
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 19.003 2.136 8.895 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 137.2 on 11 Df
Pseudo R-squared: 0.116
Number of iterations: 18 (BFGS) + 2 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + target, data = d_sim_th_children)
Residuals:
Min 1Q Median 3Q Max
-1.7560 -0.6538 -0.2126 0.5222 2.9321
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.005421 0.079310 0.068 0.9456
scale(age) -0.103910 0.079957 -1.300 0.1959
gender_m 0.068863 0.080833 0.852 0.3957
target1 -0.047832 0.239150 -0.200 0.8418
target2 -0.237626 0.231949 -1.024 0.3074
target3 -0.010505 0.238525 -0.044 0.9649
target4 -0.283076 0.238106 -1.189 0.2365
target5 -0.333051 0.238144 -1.399 0.1642
target6 -0.094453 0.238499 -0.396 0.6927
target7 -0.389305 0.238023 -1.636 0.1042
target8 0.458324 0.237920 1.926 0.0561 .
target9 0.515521 0.232123 2.221 0.0280 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9715 on 140 degrees of freedom
Multiple R-squared: 0.1249, Adjusted R-squared: 0.05617
F-statistic: 1.817 on 11 and 140 DF, p-value: 0.05638Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + +(1 | target)
Data: d_sim_th_children
REML criterion at convergence: 435.1
Scaled residuals:
Min 1Q Median 3Q Max
-1.5498 -0.7577 -0.2870 0.6243 3.0218
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.05678 0.2383
Residual 0.94357 0.9714
Number of obs: 152, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 0.007629 0.109367 9.128123 0.070 0.946
scale(age) -0.103718 0.079594 142.376071 -1.303 0.195
gender_m 0.080764 0.080103 144.424864 1.008 0.315
Correlation of Fixed Effects:
(Intr) scl(g)
scale(age) -0.005
gender_m 0.078 -0.072
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + target, data = d_sim_th_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.1108 -0.7106 -0.1652 0.6362 2.8274
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.93763 0.04040 -23.206 <2e-16 ***
scale(age) -0.05734 0.04014 -1.428 0.1532
gender_m 0.02530 0.04055 0.624 0.5327
target1 -0.01090 0.11989 -0.091 0.9276
target2 -0.09795 0.11838 -0.827 0.4080
target3 -0.01588 0.12007 -0.132 0.8948
target4 -0.16828 0.12307 -1.367 0.1715
target5 -0.15906 0.12317 -1.291 0.1966
target6 -0.04316 0.12067 -0.358 0.7206
target7 -0.23011 0.12479 -1.844 0.0652 .
target8 0.23550 0.11528 2.043 0.0411 *
target9 0.24164 0.11245 2.149 0.0316 *
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 19.303 2.171 8.892 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 138.4 on 13 Df
Pseudo R-squared: 0.1297
Number of iterations: 22 (BFGS) + 1 (Fisher scoring) China
China adults
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_ch_adults)
Residuals:
Min 1Q Median 3Q Max
-1.98225 -0.96533 -0.08779 0.84375 1.81946
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.087e-16 8.600e-02 0.000 1.000
scale(age) -4.124e-02 8.631e-02 -0.478 0.634
Residual standard error: 1.003 on 134 degrees of freedom
Multiple R-squared: 0.001701, Adjusted R-squared: -0.005749
F-statistic: 0.2283 on 1 and 134 DF, p-value: 0.6336
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_ch_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.3855 -0.8880 0.0707 0.8367 1.5746
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.81782 0.05788 -14.131 <2e-16 ***
scale(age) -0.02482 0.05701 -0.435 0.663
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 9.263 1.078 8.593 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 76.02 on 3 Df
Pseudo R-squared: 0.001559
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_ch_adults)
Residuals:
Min 1Q Median 3Q Max
-2.1027 -0.9249 -0.1069 0.8721 1.8977
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.01600 0.08581 0.186 0.852
gender_m 0.13607 0.08581 1.586 0.115
Residual standard error: 0.9948 on 133 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.01855, Adjusted R-squared: 0.01117
F-statistic: 2.514 on 1 and 133 DF, p-value: 0.1152
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_ch_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.6655 -0.8361 0.0343 0.8627 1.6182
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.80970 0.05793 -13.978 <2e-16 ***
gender_m 0.06867 0.05682 1.209 0.227
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 9.33 1.09 8.56 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 75.8 on 3 Df
Pseudo R-squared: 0.0105
Number of iterations: 10 (BFGS) + 2 (Fisher scoring) Race/ethnicity
Education
Call:
lm(formula = scale(MSE) ~ scale(education_catX), data = d_sim_ch_adults %>%
mutate(education_catX = as.numeric(education_catX)))
Residuals:
Min 1Q Median 3Q Max
-1.83531 -0.97463 -0.09368 0.86909 1.90642
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 2.856e-05 8.528e-02 0.000 1.00
scale(education_catX) -1.377e-01 8.560e-02 -1.608 0.11
Residual standard error: 0.9871 on 132 degrees of freedom
(2 observations deleted due to missingness)
Multiple R-squared: 0.01922, Adjusted R-squared: 0.01179
F-statistic: 2.587 on 1 and 132 DF, p-value: 0.1102
Call:
betareg(formula = MSE_rescaled ~ scale(education_catX), data = d_sim_ch_adults %>%
mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.3273 -0.9174 0.0886 0.8640 1.6836
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.81960 0.05741 -14.277 <2e-16 ***
scale(education_catX) -0.10514 0.05630 -1.867 0.0619 .
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 9.622 1.130 8.517 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 77.22 on 3 Df
Pseudo R-squared: 0.0242
Number of iterations: 13 (BFGS) + 1 (Fisher scoring) 
Call:
lm(formula = scale(MSE) ~ education_cat2, data = d_sim_ch_adults)
Residuals:
Min 1Q Median 3Q Max
-1.80302 -0.94233 -0.08729 0.89331 1.93871
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.02294 0.08562 0.268 0.7891
education_cat2_coll -0.17058 0.08562 -1.992 0.0484 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9821 on 132 degrees of freedom
(2 observations deleted due to missingness)
Multiple R-squared: 0.0292, Adjusted R-squared: 0.02184
F-statistic: 3.97 on 1 and 132 DF, p-value: 0.04839
Call:
betareg(formula = MSE_rescaled ~ education_cat2, data = d_sim_ch_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.2946 -0.8818 0.0835 0.8620 1.7224
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.80308 0.05727 -14.022 <2e-16 ***
education_cat2_coll -0.13018 0.05627 -2.313 0.0207 *
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 9.755 1.146 8.512 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 78.14 on 3 Df
Pseudo R-squared: 0.0373
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) Rural/urban
Call:
lm(formula = scale(MSE) ~ urban_rural_cat2, data = d_sim_ch_adults)
Residuals:
Min 1Q Median 3Q Max
-1.9687 -1.0190 -0.1139 0.8448 1.8166
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.003714 0.087560 -0.042 0.966
urban_rural_cat2_rural -0.021766 0.087560 -0.249 0.804
Residual standard error: 1.007 on 133 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.0004644, Adjusted R-squared: -0.007051
F-statistic: 0.06179 on 1 and 133 DF, p-value: 0.8041
Call:
betareg(formula = MSE_rescaled ~ urban_rural_cat2, data = d_sim_ch_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.4549 -0.9255 0.0585 0.8343 1.5591
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.818885 0.058878 -13.908 <2e-16 ***
urban_rural_cat2_rural -0.005068 0.057716 -0.088 0.93
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 9.188 1.073 8.564 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 75.02 on 3 Df
Pseudo R-squared: 5.639e-05
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) Religion
Call:
lm(formula = scale(MSE) ~ religion_cat3, data = d_sim_ch_adults)
Residuals:
Min 1Q Median 3Q Max
-1.92818 -0.94171 -0.04716 0.81335 1.78728
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.1581 0.1378 1.147 0.254
religion_cat3_buddhist -0.1663 0.1883 -0.883 0.379
religion_cat3_other 0.3205 0.2357 1.360 0.177
Residual standard error: 0.994 on 111 degrees of freedom
(22 observations deleted due to missingness)
Multiple R-squared: 0.01693, Adjusted R-squared: -0.0007785
F-statistic: 0.9561 on 2 and 111 DF, p-value: 0.3876
Call:
betareg(formula = MSE_rescaled ~ religion_cat3, data = d_sim_ch_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.4106 -0.8767 0.0854 0.8348 1.5985
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.71395 0.08789 -8.123 4.55e-16 ***
religion_cat3_buddhist -0.08271 0.12047 -0.687 0.492
religion_cat3_other 0.18629 0.14823 1.257 0.209
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 9.831 1.251 7.856 3.98e-15 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 65.74 on 4 Df
Pseudo R-squared: 0.01404
Number of iterations: 10 (BFGS) + 2 (Fisher scoring) Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_ch_adults, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-1.8115 -0.7021 -0.1410 0.8369 1.8917
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.001568 0.082752 0.019 0.98491
target1 -0.389446 0.244940 -1.590 0.11435
target2 -0.044120 0.244940 -0.180 0.85734
target3 0.129110 0.244940 0.527 0.59904
target4 0.489456 0.253148 1.933 0.05542 .
target5 -0.327869 0.244940 -1.339 0.18312
target6 0.020732 0.244940 0.085 0.93268
target7 -0.739605 0.253148 -2.922 0.00413 **
target8 0.179084 0.253148 0.707 0.48061
target9 0.398358 0.244940 1.626 0.10637
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9644 on 126 degrees of freedom
Multiple R-squared: 0.1319, Adjusted R-squared: 0.06992
F-statistic: 2.128 on 9 and 126 DF, p-value: 0.03173
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_ch_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.9833 -0.7049 -0.0544 0.8892 1.9400
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.82754 0.05447 -15.192 < 2e-16 ***
target1 -0.25057 0.16347 -1.533 0.12532
target2 0.01618 0.15710 0.103 0.91799
target3 0.07734 0.15588 0.496 0.61979
target4 0.34541 0.15673 2.204 0.02753 *
target5 -0.35136 0.16632 -2.113 0.03464 *
target6 0.03774 0.15666 0.241 0.80962
target7 -0.47680 0.17598 -2.709 0.00674 **
target8 0.15236 0.15967 0.954 0.33998
target9 0.25788 0.15287 1.687 0.09161 .
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 10.828 1.268 8.538 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 86.48 on 11 Df
Pseudo R-squared: 0.1445
Number of iterations: 21 (BFGS) + 2 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + scale(education_catX) +
religion_cat3 + urban_rural_cat2 + target, data = d_sim_ch_adults %>%
mutate(education_catX = as.numeric(education_catX)))
Residuals:
Min 1Q Median 3Q Max
-1.80819 -0.71922 -0.04958 0.84399 1.84406
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.10191 0.14769 0.690 0.492
scale(age) 0.02301 0.11639 0.198 0.844
gender_m 0.10885 0.10180 1.069 0.288
scale(education_catX) -0.15335 0.11667 -1.314 0.192
religion_cat3_buddhist -0.04976 0.19828 -0.251 0.802
religion_cat3_other 0.15413 0.25251 0.610 0.543
urban_rural_cat2_rural -0.08875 0.10681 -0.831 0.408
target1 -0.27630 0.27711 -0.997 0.321
target2 -0.36463 0.33601 -1.085 0.281
target3 0.08036 0.25961 0.310 0.758
target4 0.52215 0.28595 1.826 0.071 .
target5 -0.23257 0.29221 -0.796 0.428
target6 -0.09449 0.26274 -0.360 0.720
target7 -0.45242 0.31609 -1.431 0.156
target8 0.24088 0.27284 0.883 0.380
target9 0.29498 0.29328 1.006 0.317
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9853 on 95 degrees of freedom
(25 observations deleted due to missingness)
Multiple R-squared: 0.1359, Adjusted R-squared: -0.0004857
F-statistic: 0.9964 on 15 and 95 DF, p-value: 0.4655
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + education_cat2 +
religion_cat3 + urban_rural_cat2 + target, data = d_sim_ch_adults)
Residuals:
Min 1Q Median 3Q Max
-1.8763 -0.7073 -0.1275 0.8295 1.8547
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.12959 0.14389 0.901 0.3701
scale(age) 0.01135 0.11545 0.098 0.9219
gender_m 0.10224 0.10128 1.009 0.3153
education_cat2_coll -0.18022 0.11307 -1.594 0.1143
religion_cat3_buddhist -0.05008 0.19692 -0.254 0.7998
religion_cat3_other 0.15839 0.24911 0.636 0.5264
urban_rural_cat2_rural -0.09263 0.10564 -0.877 0.3828
target1 -0.27895 0.27551 -1.012 0.3139
target2 -0.35094 0.33329 -1.053 0.2950
target3 0.06198 0.25926 0.239 0.8116
target4 0.53953 0.28512 1.892 0.0615 .
target5 -0.22798 0.29054 -0.785 0.4346
target6 -0.07104 0.26096 -0.272 0.7860
target7 -0.46661 0.31462 -1.483 0.1414
target8 0.22954 0.27000 0.850 0.3974
target9 0.26820 0.29235 0.917 0.3613
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9812 on 95 degrees of freedom
(25 observations deleted due to missingness)
Multiple R-squared: 0.1431, Adjusted R-squared: 0.007854
F-statistic: 1.058 on 15 and 95 DF, p-value: 0.4054Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + scale(education_catX) + religion_cat3 +
urban_rural_cat2 + +(1 | target)
Data: d_sim_ch_adults %>% mutate(education_catX = as.numeric(education_catX))
REML criterion at convergence: 322.8
Scaled residuals:
Min 1Q Median 3Q Max
-2.02986 -0.84997 -0.07919 0.79709 1.82275
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.001236 0.03516
Residual 0.976187 0.98802
Number of obs: 111, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 0.156971 0.142408 26.696548 1.102 0.280
scale(age) -0.007222 0.113264 102.272384 -0.064 0.949
gender_m 0.131768 0.094980 97.398767 1.387 0.169
scale(education_catX) -0.128565 0.114095 101.871942 -1.127 0.262
religion_cat3_buddhist -0.106878 0.195618 99.246537 -0.546 0.586
religion_cat3_other 0.260819 0.242232 103.986230 1.077 0.284
urban_rural_cat2_rural -0.046704 0.101603 103.886042 -0.460 0.647
Correlation of Fixed Effects:
(Intr) scl(g) gndr_m sc(_X) rlgn_ct3_b rlgn_ct3_t
scale(age) 0.067
gender_m 0.033 0.117
scl(dctn_X) 0.180 0.463 -0.002
rlgn_ct3_bd -0.105 -0.194 -0.021 -0.113
rlgn_ct3_th 0.547 0.151 0.019 0.181 -0.772
urbn_rrl_2_ 0.176 -0.087 -0.083 0.249 -0.110 0.158 Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + education_cat2 + religion_cat3 +
urban_rural_cat2 + +(1 | target)
Data: d_sim_ch_adults
REML criterion at convergence: 322
Scaled residuals:
Min 1Q Median 3Q Max
-2.0241 -0.8635 -0.0846 0.8195 1.8610
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.001573 0.03966
Residual 0.968531 0.98414
Number of obs: 111, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 0.17841 0.13976 26.03653 1.277 0.213
scale(age) -0.01944 0.11166 103.17213 -0.174 0.862
gender_m 0.12745 0.09468 97.79777 1.346 0.181
education_cat2_coll -0.16033 0.11144 100.32917 -1.439 0.153
religion_cat3_buddhist -0.10567 0.19445 99.01632 -0.543 0.588
religion_cat3_other 0.26022 0.23983 103.88604 1.085 0.280
urban_rural_cat2_rural -0.05277 0.10092 103.99192 -0.523 0.602
Correlation of Fixed Effects:
(Intr) scl(g) gndr_m edc_2_ rlgn_ct3_b rlgn_ct3_t
scale(age) -0.001
gender_m 0.035 0.132
edctn_ct2_c 0.037 0.445 0.030
rlgn_ct3_bd -0.089 -0.184 -0.024 -0.093
rlgn_ct3_th 0.530 0.133 0.024 0.144 -0.772
urbn_rrl_2_ 0.142 -0.099 -0.076 0.238 -0.104 0.149
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + scale(education_catX) + religion_cat3 +
urban_rural_cat2 + target, data = d_sim_ch_adults %>% mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.2342 -0.7313 0.0402 0.9603 1.9925
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.76244 0.08999 -8.473 <2e-16 ***
scale(age) 0.03992 0.07115 0.561 0.5748
gender_m 0.06672 0.06240 1.069 0.2849
scale(education_catX) -0.09461 0.07193 -1.315 0.1884
religion_cat3_buddhist -0.02384 0.12011 -0.199 0.8427
religion_cat3_other 0.08275 0.15112 0.548 0.5840
urban_rural_cat2_rural -0.05144 0.06600 -0.779 0.4358
target1 -0.18546 0.17421 -1.065 0.2871
target2 -0.20030 0.20975 -0.955 0.3396
target3 0.04251 0.15848 0.268 0.7885
target4 0.35641 0.16940 2.104 0.0354 *
target5 -0.25484 0.18546 -1.374 0.1694
target6 -0.03240 0.16074 -0.202 0.8403
target7 -0.25495 0.20345 -1.253 0.2102
target8 0.17183 0.16549 1.038 0.2991
target9 0.18358 0.17598 1.043 0.2969
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 11.398 1.479 7.709 1.27e-14 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 71.79 on 17 Df
Pseudo R-squared: 0.1356
Number of iterations: 26 (BFGS) + 2 (Fisher scoring)
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + education_cat2 + religion_cat3 +
urban_rural_cat2 + target, data = d_sim_ch_adults %>% mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.2420 -0.7853 0.0128 0.9615 1.9850
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.74582 0.08748 -8.526 <2e-16 ***
scale(age) 0.03074 0.07061 0.435 0.6633
gender_m 0.06228 0.06211 1.003 0.3160
education_cat2_coll -0.11724 0.06958 -1.685 0.0920 .
religion_cat3_buddhist -0.02138 0.11922 -0.179 0.8577
religion_cat3_other 0.08214 0.14904 0.551 0.5815
urban_rural_cat2_rural -0.05626 0.06533 -0.861 0.3891
target1 -0.18743 0.17317 -1.082 0.2791
target2 -0.19209 0.20794 -0.924 0.3556
target3 0.02920 0.15839 0.184 0.8537
target4 0.36565 0.16895 2.164 0.0304 *
target5 -0.24980 0.18436 -1.355 0.1754
target6 -0.02218 0.15974 -0.139 0.8896
target7 -0.26495 0.20257 -1.308 0.1909
target8 0.16606 0.16376 1.014 0.3106
target9 0.16899 0.17536 0.964 0.3352
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 11.516 1.494 7.706 1.3e-14 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 72.37 on 17 Df
Pseudo R-squared: 0.1453
Number of iterations: 26 (BFGS) + 2 (Fisher scoring) China children
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_ch_children)
Residuals:
Min 1Q Median 3Q Max
-1.7635 -0.7910 -0.1152 0.7655 2.5044
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.0007363 0.0876920 0.008 0.993
scale(age) 0.0902228 0.0880741 1.024 0.308
Residual standard error: 0.9998 on 128 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.008132, Adjusted R-squared: 0.0003827
F-statistic: 1.049 on 1 and 128 DF, p-value: 0.3076
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_ch_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.0806 -0.6844 0.0900 0.8033 1.9425
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.18297 0.06077 -19.465 <2e-16 ***
scale(age) 0.05430 0.05864 0.926 0.354
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 10.545 1.274 8.279 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 94.81 on 3 Df
Pseudo R-squared: 0.007183
Number of iterations: 13 (BFGS) + 1 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_ch_children)
Residuals:
Min 1Q Median 3Q Max
-1.63203 -0.85381 -0.08124 0.71790 2.46064
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.0005309 0.0880053 -0.006 0.995
gender_m -0.0345062 0.0880053 -0.392 0.696
Residual standard error: 1.003 on 128 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.0012, Adjusted R-squared: -0.006604
F-statistic: 0.1537 on 1 and 128 DF, p-value: 0.6956
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_ch_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.8828 -0.7265 0.1241 0.7659 1.9251
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.18303 0.06095 -19.411 <2e-16 ***
gender_m -0.01553 0.05860 -0.265 0.791
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 10.478 1.265 8.281 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 94.4 on 3 Df
Pseudo R-squared: 0.0005475
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_ch_children, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-1.94725 -0.55250 -0.07091 0.60449 2.59776
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.05087 0.11494 0.443 0.65887
target1 -0.02600 0.26410 -0.098 0.92173
target2 0.15149 0.25634 0.591 0.55566
target3 0.58204 0.86498 0.673 0.50232
target4 -0.20095 0.28289 -0.710 0.47888
target5 -0.06192 0.25634 -0.242 0.80953
target6 0.37729 0.26410 1.429 0.15575
target7 -0.50273 0.27287 -1.842 0.06791 .
target8 -0.80103 0.26410 -3.033 0.00297 **
target9 0.10309 0.25634 0.402 0.68830
target10 -0.34742 0.26410 -1.315 0.19087
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9478 on 119 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.1713, Adjusted R-squared: 0.1017
F-statistic: 2.46 on 10 and 119 DF, p-value: 0.0104
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_ch_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.0700 -0.5880 0.0966 0.7542 2.2198
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.16378 0.07347 -15.840 < 2e-16 ***
target1 -0.02523 0.17098 -0.148 0.88269
target2 0.15964 0.16118 0.990 0.32194
target3 0.44977 0.51959 0.866 0.38670
target4 -0.15569 0.18768 -0.830 0.40678
target5 -0.02180 0.16579 -0.131 0.89540
target6 0.17170 0.16577 1.036 0.30031
target7 -0.33839 0.18735 -1.806 0.07089 .
target8 -0.59175 0.19075 -3.102 0.00192 **
target9 0.11634 0.16221 0.717 0.47324
target10 -0.21760 0.17695 -1.230 0.21880
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 12.649 1.536 8.235 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 106.3 on 12 Df
Pseudo R-squared: 0.1726
Number of iterations: 21 (BFGS) + 2 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + target, data = d_sim_ch_children)
Residuals:
Min 1Q Median 3Q Max
-1.78651 -0.59622 -0.05927 0.52989 2.55854
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.037040 0.115840 0.320 0.74972
scale(age) 0.064963 0.088003 0.738 0.46187
gender_m -0.087524 0.086622 -1.010 0.31438
target1 0.008039 0.266616 0.030 0.97600
target2 0.134658 0.259306 0.519 0.60453
target3 0.439427 0.876055 0.502 0.61689
target4 -0.141745 0.288276 -0.492 0.62385
target5 -0.015492 0.260035 -0.060 0.95260
target6 0.381900 0.265512 1.438 0.15300
target7 -0.529608 0.274554 -1.929 0.05616 .
target8 -0.805686 0.265134 -3.039 0.00293 **
target9 0.147945 0.260197 0.569 0.57072
target10 -0.347887 0.267575 -1.300 0.19611
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9502 on 117 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.181, Adjusted R-squared: 0.09704
F-statistic: 2.155 on 12 and 117 DF, p-value: 0.01829Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + +(1 | target)
Data: d_sim_ch_children
REML criterion at convergence: 370.4
Scaled residuals:
Min 1Q Median 3Q Max
-1.6938 -0.7400 -0.1060 0.6657 2.5110
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.1183 0.3439
Residual 0.8989 0.9481
Number of obs: 130, groups: target, 11
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 0.003062 0.135961 9.250871 0.023 0.983
scale(age) 0.079035 0.086095 123.704921 0.918 0.360
gender_m -0.073540 0.085141 122.345763 -0.864 0.389
Correlation of Fixed Effects:
(Intr) scl(g)
scale(age) -0.003
gender_m 0.017 -0.122Vanuatu
Vanuatu adults
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_vt_adults)
Residuals:
Min 1Q Median 3Q Max
-1.3978 -1.0958 -0.1866 0.9466 1.6399
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.005177 0.084151 -0.062 0.951
scale(age) 0.004487 0.084449 0.053 0.958
Residual standard error: 1.003 on 140 degrees of freedom
(6 observations deleted due to missingness)
Multiple R-squared: 2.017e-05, Adjusted R-squared: -0.007123
F-statistic: 0.002824 on 1 and 140 DF, p-value: 0.9577
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_vt_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.5191 -1.1249 0.1113 0.8946 1.3277
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.101665 0.075570 -14.58 <2e-16 ***
scale(age) -0.007842 0.071174 -0.11 0.912
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 5.4708 0.6208 8.812 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 69.24 on 3 Df
Pseudo R-squared: 0.000101
Number of iterations: 13 (BFGS) + 1 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_vt_adults)
Residuals:
Min 1Q Median 3Q Max
-1.4115 -1.0972 -0.1551 0.9793 1.6432
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.01622 0.08903 0.182 0.856
gender_m 0.04287 0.08903 0.482 0.631
Residual standard error: 1.003 on 146 degrees of freedom
Multiple R-squared: 0.001586, Adjusted R-squared: -0.005253
F-statistic: 0.2319 on 1 and 146 DF, p-value: 0.6309
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_vt_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.4878 -1.1435 0.1294 0.9184 1.3132
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.08823 0.07900 -13.774 <2e-16 ***
gender_m 0.02521 0.07481 0.337 0.736
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 5.4614 0.6069 8.999 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 71.84 on 3 Df
Pseudo R-squared: 0.0007893
Number of iterations: 13 (BFGS) + 2 (Fisher scoring) Location
Call:
lm(formula = scale(MSE) ~ location_cat2, data = d_sim_vt_adults)
Residuals:
Min 1Q Median 3Q Max
-1.57886 -0.92190 -0.09571 0.96541 1.83714
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -2.053e-17 8.032e-02 0.000 1.00000
location_cat2_urban 2.264e-01 8.032e-02 2.819 0.00549 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9772 on 146 degrees of freedom
Multiple R-squared: 0.05162, Adjusted R-squared: 0.04512
F-statistic: 7.946 on 1 and 146 DF, p-value: 0.005488
Call:
betareg(formula = MSE_rescaled ~ location_cat2, data = d_sim_vt_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.4571 -1.0025 0.1772 0.9292 1.4307
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.10317 0.07311 -15.089 <2e-16 ***
location_cat2_urban 0.16126 0.06878 2.345 0.0191 *
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 5.6687 0.6314 8.978 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 74.49 on 3 Df
Pseudo R-squared: 0.03722
Number of iterations: 14 (BFGS) + 2 (Fisher scoring) Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_vt_adults, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-2.1363 -0.6290 -0.1473 0.6257 2.0004
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.03428 0.08934 0.384 0.7018
target1 -0.43628 0.22886 -1.906 0.0587 .
target2 0.18645 0.22886 0.815 0.4167
target3 -0.03812 0.24333 -0.157 0.8758
target4 0.35781 0.58392 0.613 0.5410
target5 0.19330 0.23569 0.820 0.4136
target6 -0.15988 0.22886 -0.699 0.4860
target7 -0.29295 0.22886 -1.280 0.2027
target8 -0.99274 0.22886 -4.338 2.77e-05 ***
target9 0.30521 0.23569 1.295 0.1975
target10 -0.07647 0.22886 -0.334 0.7388
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9022 on 137 degrees of freedom
Multiple R-squared: 0.2414, Adjusted R-squared: 0.186
F-statistic: 4.36 on 10 and 137 DF, p-value: 2.616e-05
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_vt_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.9923 -0.5070 0.1645 0.7768 1.7255
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.09169 0.07850 -13.906 < 2e-16 ***
target1 -0.38694 0.20656 -1.873 0.061030 .
target2 0.08989 0.19358 0.464 0.642375
target3 -0.04871 0.20950 -0.233 0.816143
target4 0.43429 0.47368 0.917 0.359216
target5 0.12096 0.19862 0.609 0.542509
target6 -0.19922 0.20103 -0.991 0.321683
target7 -0.15220 0.19972 -0.762 0.446023
target8 -0.83086 0.22069 -3.765 0.000167 ***
target9 0.26711 0.19544 1.367 0.171702
target10 -0.04615 0.19690 -0.234 0.814684
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 6.9512 0.7828 8.88 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 89.01 on 12 Df
Pseudo R-squared: 0.2161
Number of iterations: 22 (BFGS) + 2 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + location_cat2 +
target, data = d_sim_vt_adults)
Residuals:
Min 1Q Median 3Q Max
-2.30230 -0.60303 -0.08047 0.59366 1.94575
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.076090 0.095069 0.800 0.42498
scale(age) -0.009992 0.075740 -0.132 0.89526
gender_m 0.069748 0.082202 0.848 0.39774
location_cat2_urban 0.208243 0.075175 2.770 0.00644 **
target1 -0.486842 0.226481 -2.150 0.03347 *
target2 0.223195 0.232775 0.959 0.33944
target3 -0.059356 0.239496 -0.248 0.80466
target4 0.604214 0.582132 1.038 0.30126
target5 0.170536 0.231806 0.736 0.46327
target6 -0.312368 0.238869 -1.308 0.19332
target7 -0.270313 0.226255 -1.195 0.23440
target8 -0.973820 0.232840 -4.182 5.31e-05 ***
target9 0.314015 0.231381 1.357 0.17713
target10 -0.141242 0.234133 -0.603 0.54740
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8832 on 128 degrees of freedom
(6 observations deleted due to missingness)
Multiple R-squared: 0.2908, Adjusted R-squared: 0.2187
F-statistic: 4.036 on 13 and 128 DF, p-value: 1.542e-05Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + location_cat2 + +(1 | target)
Data: d_sim_vt_adults
REML criterion at convergence: 391.1
Scaled residuals:
Min 1Q Median 3Q Max
-2.3711 -0.7719 -0.1157 0.7591 2.1208
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.2061 0.4540
Residual 0.7789 0.8826
Number of obs: 142, groups: target, 11
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 3.893e-02 1.613e-01 1.046e+01 0.241 0.81394
scale(age) -2.487e-04 7.522e-02 1.302e+02 -0.003 0.99737
gender_m 5.942e-02 8.140e-02 1.321e+02 0.730 0.46668
location_cat2_urban 2.055e-01 7.476e-02 1.299e+02 2.750 0.00682 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Correlation of Fixed Effects:
(Intr) scl(g) gndr_m
scale(age) -0.024
gender_m 0.188 -0.065
lctn_ct2_rb 0.013 0.067 -0.032
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + location_cat2 + target, data = d_sim_vt_adults)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.5871 -0.5685 0.1536 0.7918 1.7780
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.06328 0.08271 -12.855 < 2e-16 ***
scale(age) -0.01158 0.06568 -0.176 0.860019
gender_m 0.06627 0.07087 0.935 0.349699
location_cat2_urban 0.16303 0.06526 2.498 0.012484 *
target1 -0.46056 0.20559 -2.240 0.025078 *
target2 0.14424 0.19473 0.741 0.458857
target3 -0.06243 0.20630 -0.303 0.762164
target4 0.63952 0.47304 1.352 0.176396
target5 0.11098 0.19521 0.569 0.569691
target6 -0.32783 0.21468 -1.527 0.126748
target7 -0.13385 0.19777 -0.677 0.498523
target8 -0.83928 0.22575 -3.718 0.000201 ***
target9 0.29020 0.19160 1.515 0.129885
target10 -0.09398 0.20161 -0.466 0.641100
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 7.3748 0.8505 8.671 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 89.52 on 15 Df
Pseudo R-squared: 0.2536
Number of iterations: 24 (BFGS) + 2 (Fisher scoring) Vanuatu children
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_vt_children)
Residuals:
Min 1Q Median 3Q Max
-1.8608 -0.9267 0.1151 0.8844 1.4462
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.01157 0.08368 0.138 0.890
scale(age) 0.02228 0.08398 0.265 0.791
Residual standard error: 0.9972 on 140 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.0005025, Adjusted R-squared: -0.006637
F-statistic: 0.07038 on 1 and 140 DF, p-value: 0.7912
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_vt_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.1805 -0.8564 0.1844 0.8695 1.3754
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.430248 0.048815 -8.814 <2e-16 ***
scale(age) 0.006664 0.048773 0.137 0.891
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 11.32 1.29 8.771 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 80.22 on 3 Df
Pseudo R-squared: 0.0001346
Number of iterations: 9 (BFGS) + 2 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_vt_children)
Residuals:
Min 1Q Median 3Q Max
-1.86647 -0.90478 0.09592 0.88104 1.46081
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.008161 0.084711 -0.096 0.923
gender_m -0.055574 0.084711 -0.656 0.513
Residual standard error: 1.002 on 141 degrees of freedom
Multiple R-squared: 0.003043, Adjusted R-squared: -0.004028
F-statistic: 0.4304 on 1 and 141 DF, p-value: 0.5129
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_vt_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.1830 -0.8390 0.1678 0.8723 1.3920
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.44208 0.04950 -8.931 <2e-16 ***
gender_m -0.03318 0.04928 -0.673 0.501
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 11.198 1.272 8.805 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 80.26 on 3 Df
Pseudo R-squared: 0.003177
Number of iterations: 11 (BFGS) + 2 (Fisher scoring) Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_vt_children, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-2.22143 -0.59576 0.05297 0.63392 2.01145
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.02832 0.07872 0.360 0.71964
target1 -0.67395 0.23952 -2.814 0.00565 **
target2 -0.51665 0.23176 -2.229 0.02749 *
target3 0.29425 0.29892 0.984 0.32672
target4 0.38686 0.34215 1.131 0.26025
target5 0.23323 0.22483 1.037 0.30145
target6 0.36705 0.23176 1.584 0.11565
target7 0.55082 0.21857 2.520 0.01293 *
target8 -0.95911 0.22483 -4.266 3.77e-05 ***
target9 -0.13908 0.22483 -0.619 0.53723
target10 0.09963 0.23952 0.416 0.67810
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9017 on 132 degrees of freedom
Multiple R-squared: 0.2442, Adjusted R-squared: 0.1869
F-statistic: 4.265 on 10 and 132 DF, p-value: 3.732e-05
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_vt_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.0861 -0.6609 0.1370 0.7399 2.2219
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.42754 0.04514 -9.472 < 2e-16 ***
target1 -0.42361 0.14315 -2.959 0.00309 **
target2 -0.31615 0.13636 -2.318 0.02043 *
target3 0.19121 0.16786 1.139 0.25467
target4 0.24399 0.19166 1.273 0.20300
target5 0.13825 0.12671 1.091 0.27527
target6 0.21419 0.13015 1.646 0.09981 .
target7 0.31819 0.12241 2.599 0.00934 **
target8 -0.56957 0.13749 -4.143 3.43e-05 ***
target9 -0.06163 0.12854 -0.479 0.63159
target10 0.05224 0.13569 0.385 0.70023
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 14.787 1.696 8.716 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 100 on 12 Df
Pseudo R-squared: 0.2479
Number of iterations: 21 (BFGS) + 2 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + target, data = d_sim_vt_children)
Residuals:
Min 1Q Median 3Q Max
-2.23284 -0.57970 0.03138 0.58494 1.98058
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.03605 0.07853 0.459 0.64698
scale(age) -0.00728 0.07643 -0.095 0.92427
gender_m -0.04021 0.07775 -0.517 0.60588
target1 -0.68118 0.23828 -2.859 0.00496 **
target2 -0.51908 0.22956 -2.261 0.02542 *
target3 0.27485 0.29517 0.931 0.35352
target4 0.38024 0.33757 1.126 0.26209
target5 0.22451 0.22239 1.010 0.31461
target6 0.34737 0.22874 1.519 0.13130
target7 0.67889 0.22244 3.052 0.00276 **
target8 -0.98538 0.22306 -4.418 2.09e-05 ***
target9 -0.16564 0.22303 -0.743 0.45903
target10 0.10364 0.23916 0.433 0.66548
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8885 on 129 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.2688, Adjusted R-squared: 0.2008
F-statistic: 3.953 on 12 and 129 DF, p-value: 3.458e-05Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + +(1 | target)
Data: d_sim_vt_children
REML criterion at convergence: 391.1
Scaled residuals:
Min 1Q Median 3Q Max
-2.44409 -0.81803 0.08568 0.71906 2.05346
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.2182 0.4672
Residual 0.7880 0.8877
Number of obs: 142, groups: target, 11
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 0.025394 0.160746 10.417215 0.158 0.877
scale(age) -0.000469 0.076012 131.630258 -0.006 0.995
gender_m -0.041024 0.077200 132.360377 -0.531 0.596
Correlation of Fixed Effects:
(Intr) scl(g)
scale(age) -0.001
gender_m 0.070 -0.035
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + target, data = d_sim_vt_children)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.2043 -0.6984 0.0942 0.7788 2.3123
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.42257 0.04466 -9.463 < 2e-16 ***
scale(age) -0.00811 0.04345 -0.187 0.85194
gender_m -0.01615 0.04421 -0.365 0.71491
target1 -0.43278 0.14123 -3.064 0.00218 **
target2 -0.32141 0.13397 -2.399 0.01644 *
target3 0.17955 0.16438 1.092 0.27470
target4 0.23957 0.18750 1.278 0.20134
target5 0.13341 0.12430 1.073 0.28316
target6 0.20318 0.12737 1.595 0.11068
target7 0.40402 0.12340 3.274 0.00106 **
target8 -0.58371 0.13530 -4.314 1.6e-05 ***
target9 -0.07637 0.12648 -0.604 0.54598
target10 0.05178 0.13439 0.385 0.70005
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 15.545 1.792 8.674 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 102.6 on 14 Df
Pseudo R-squared: 0.2747
Number of iterations: 22 (BFGS) + 2 (Fisher scoring) Using adults’ models for individual children
US
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_us_adch)
Residuals:
Min 1Q Median 3Q Max
-1.5302 -0.7290 -0.2359 0.7080 2.2975
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.02368 0.09131 0.259 0.79584
scale(age) -0.24796 0.09173 -2.703 0.00797 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9621 on 109 degrees of freedom
(6 observations deleted due to missingness)
Multiple R-squared: 0.06282, Adjusted R-squared: 0.05423
F-statistic: 7.307 on 1 and 109 DF, p-value: 0.00797
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_us_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.9037 -0.6458 -0.0220 0.8135 1.8915
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.07931 0.06948 -15.535 < 2e-16 ***
scale(age) -0.19634 0.06735 -2.915 0.00355 **
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 8.858 1.150 7.702 1.34e-14 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 70.28 on 3 Df
Pseudo R-squared: 0.07682
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_us_adch)
Residuals:
Min 1Q Median 3Q Max
-1.59916 -0.83868 -0.09822 0.64239 2.15513
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.01470 0.09337 -0.157 0.875
gender_m -0.10117 0.09337 -1.083 0.281
Residual standard error: 0.9993 on 115 degrees of freedom
Multiple R-squared: 0.0101, Adjusted R-squared: 0.001497
F-statistic: 1.174 on 1 and 115 DF, p-value: 0.2809
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_us_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.8536 -0.7803 0.1046 0.7064 1.7097
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.09860 0.07147 -15.37 <2e-16 ***
gender_m -0.05957 0.06850 -0.87 0.384
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 8.048 1.015 7.929 2.2e-15 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 69.79 on 3 Df
Pseudo R-squared: 0.006656
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) Race/ethnicity
Call:
lm(formula = scale(MSE) ~ ethnicity_cat2, data = d_sim_us_adch)
Residuals:
Min 1Q Median 3Q Max
-1.73204 -0.74203 -0.05494 0.70526 2.21985
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.01171 0.10684 -0.110 0.913
ethnicity_cat2_POC -0.15324 0.10684 -1.434 0.155
Residual standard error: 0.9608 on 90 degrees of freedom
(25 observations deleted due to missingness)
Multiple R-squared: 0.02235, Adjusted R-squared: 0.01148
F-statistic: 2.057 on 1 and 90 DF, p-value: 0.155
Call:
betareg(formula = MSE_rescaled ~ ethnicity_cat2, data = d_sim_us_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.2025 -0.6368 0.1387 0.7532 1.7655
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.11860 0.08348 -13.400 <2e-16 ***
ethnicity_cat2_POC -0.06768 0.08028 -0.843 0.399
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 8.325 1.188 7.009 2.4e-12 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 57.84 on 3 Df
Pseudo R-squared: 0.007623
Number of iterations: 8 (BFGS) + 2 (Fisher scoring) Religion
Call:
lm(formula = scale(MSE) ~ religion_cat3, data = d_sim_us_adch)
Residuals:
Min 1Q Median 3Q Max
-1.6670 -0.7996 -0.1913 0.6339 1.9324
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.07364 0.10419 -0.707 0.4813
religion_cat3_christian 0.15016 0.13389 1.122 0.2648
religion_cat3_other -0.34635 0.16113 -2.150 0.0341 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9985 on 98 degrees of freedom
(16 observations deleted due to missingness)
Multiple R-squared: 0.04503, Adjusted R-squared: 0.02554
F-statistic: 2.31 on 2 and 98 DF, p-value: 0.1046
Call:
betareg(formula = MSE_rescaled ~ religion_cat3, data = d_sim_us_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.1849 -0.7007 0.0240 0.7135 1.6105
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.14672 0.08038 -14.266 <2e-16 ***
religion_cat3_christian 0.10450 0.09796 1.067 0.2861
religion_cat3_other -0.25667 0.12180 -2.107 0.0351 *
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 8.250 1.122 7.353 1.94e-13 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 62.15 on 4 Df
Pseudo R-squared: 0.04868
Number of iterations: 11 (BFGS) + 2 (Fisher scoring) Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_us_adch, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-1.5739 -0.6128 -0.1199 0.5806 2.3814
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.021141 0.076516 -0.276 0.782854
target1 -0.565637 0.235312 -2.404 0.017946 *
target2 0.142811 0.226374 0.631 0.529477
target3 -0.003955 0.218526 -0.018 0.985595
target4 -0.189887 0.226374 -0.839 0.403441
target5 -0.265973 0.218526 -1.217 0.226237
target6 -0.373798 0.235312 -1.589 0.115120
target7 -0.950286 0.245608 -3.869 0.000188 ***
target8 0.096565 0.235312 0.410 0.682354
target9 0.884649 0.226374 3.908 0.000164 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8251 on 107 degrees of freedom
Multiple R-squared: 0.372, Adjusted R-squared: 0.3191
F-statistic: 7.041 on 9 and 107 DF, p-value: 6.066e-08
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_us_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.9222 -0.6841 0.0495 0.7834 2.4545
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.15756 0.06031 -19.192 < 2e-16 ***
target1 -0.40022 0.18988 -2.108 0.035054 *
target2 0.19694 0.16334 1.206 0.227934
target3 -0.01581 0.16338 -0.097 0.922903
target4 -0.13562 0.17302 -0.784 0.433124
target5 -0.12492 0.16668 -0.750 0.453551
target6 -0.36946 0.18864 -1.959 0.050163 .
target7 -0.78015 0.21595 -3.613 0.000303 ***
target8 0.08559 0.17277 0.495 0.620326
target9 0.66397 0.15461 4.294 1.75e-05 ***
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 12.689 1.625 7.809 5.77e-15 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 95.65 on 11 Df
Pseudo R-squared: 0.3533
Number of iterations: 19 (BFGS) + 2 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + ethnicity_cat2 +
religion_cat3 + target, data = d_sim_us_adch)
Residuals:
Min 1Q Median 3Q Max
-1.1612 -0.5556 -0.1462 0.4263 2.2624
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.0003225 0.1168310 -0.003 0.997806
scale(age) -0.2151615 0.0927814 -2.319 0.023704 *
gender_m 0.0252135 0.1058055 0.238 0.812434
ethnicity_cat2_POC -0.1467148 0.1161566 -1.263 0.211291
religion_cat3_christian 0.0101795 0.1437128 0.071 0.943759
religion_cat3_other -0.0546606 0.1817050 -0.301 0.764558
target1 -0.5212159 0.3018113 -1.727 0.089154 .
target2 0.0397639 0.2764660 0.144 0.886102
target3 0.3351109 0.2545001 1.317 0.192772
target4 0.1277681 0.3191201 0.400 0.690256
target5 -0.2934084 0.2942442 -0.997 0.322563
target6 -0.5507955 0.3071590 -1.793 0.077820 .
target7 -0.8386141 0.3146453 -2.665 0.009795 **
target8 -0.0657910 0.2915482 -0.226 0.822207
target9 1.0654590 0.2648380 4.023 0.000159 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8124 on 62 degrees of freedom
(40 observations deleted due to missingness)
Multiple R-squared: 0.4394, Adjusted R-squared: 0.3128
F-statistic: 3.471 on 14 and 62 DF, p-value: 0.00035Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + ethnicity_cat2 + religion_cat3 + +(1 | target)
Data: d_sim_us_adch
REML criterion at convergence: 208.1
Scaled residuals:
Min 1Q Median 3Q Max
-1.4287 -0.5873 -0.1834 0.4568 2.5660
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.2476 0.4976
Residual 0.6603 0.8126
Number of obs: 77, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) -0.005392 0.195523 10.964523 -0.028 0.978
scale(age) -0.228717 0.091553 65.011447 -2.498 0.015 *
gender_m -0.011599 0.102618 68.288867 -0.113 0.910
ethnicity_cat2_POC -0.159959 0.112348 68.634615 -1.424 0.159
religion_cat3_christian 0.009530 0.139824 67.870228 0.068 0.946
religion_cat3_other -0.109178 0.177292 67.345279 -0.616 0.540
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Correlation of Fixed Effects:
(Intr) scl(g) gndr_m e_2_PO rlgn_ct3_c
scale(age) -0.049
gender_m 0.042 -0.089
ethnc_2_POC -0.259 0.028 0.170
rlgn_ct3_ch -0.198 0.031 -0.124 0.068
rlgn_ct3_th 0.244 0.002 0.024 -0.117 -0.609
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + ethnicity_cat2 + religion_cat3 +
target, data = d_sim_us_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.9154 -0.7664 -0.0799 0.7831 2.7043
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.170578 0.085653 -13.667 < 2e-16 ***
scale(age) -0.197720 0.065407 -3.023 0.00250 **
gender_m -0.003992 0.076601 -0.052 0.95844
ethnicity_cat2_POC -0.107063 0.080914 -1.323 0.18578
religion_cat3_christian 0.028474 0.103690 0.275 0.78362
religion_cat3_other -0.072890 0.134131 -0.543 0.58684
target1 -0.424284 0.232128 -1.828 0.06758 .
target2 0.105307 0.192853 0.546 0.58503
target3 0.265353 0.173784 1.527 0.12678
target4 0.155570 0.225445 0.690 0.49016
target5 -0.161728 0.216744 -0.746 0.45556
target6 -0.518757 0.232888 -2.228 0.02591 *
target7 -0.732025 0.267842 -2.733 0.00628 **
target8 -0.026639 0.209091 -0.127 0.89862
target9 0.789917 0.171210 4.614 3.96e-06 ***
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 15.155 2.405 6.302 2.93e-10 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 70.54 on 16 Df
Pseudo R-squared: 0.4389
Number of iterations: 25 (BFGS) + 1 (Fisher scoring) Ghana
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_gh_adch)
Residuals:
Min 1Q Median 3Q Max
-1.5741 -0.9666 0.1471 0.9985 1.4829
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 2.665e-16 8.175e-02 0.000 1.000
scale(age) 6.578e-02 8.202e-02 0.802 0.424
Residual standard error: 1.001 on 148 degrees of freedom
Multiple R-squared: 0.004326, Adjusted R-squared: -0.002401
F-statistic: 0.6431 on 1 and 148 DF, p-value: 0.4239
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_gh_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.0877 -0.7355 0.3249 0.9140 1.2440
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.89197 0.07099 -12.565 <2e-16 ***
scale(age) 0.06120 0.06836 0.895 0.371
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 5.2985 0.5764 9.193 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 58.44 on 3 Df
Pseudo R-squared: 0.005533
Number of iterations: 15 (BFGS) + 2 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_gh_adch)
Residuals:
Min 1Q Median 3Q Max
-1.4895 -0.9795 0.1744 1.0006 1.4029
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.0006425 0.0820358 -0.008 0.994
gender_m -0.0120460 0.0820358 -0.147 0.883
Residual standard error: 1.003 on 148 degrees of freedom
Multiple R-squared: 0.0001457, Adjusted R-squared: -0.00661
F-statistic: 0.02156 on 1 and 148 DF, p-value: 0.8835
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_gh_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.9453 -0.7459 0.3580 0.9217 1.1737
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.89251 0.07123 -12.531 <2e-16 ***
gender_m -0.01989 0.06826 -0.291 0.771
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 5.2719 0.5733 9.196 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 58.07 on 3 Df
Pseudo R-squared: 0.0005525
Number of iterations: 13 (BFGS) + 2 (Fisher scoring) Religion
Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_gh_adch, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-1.87020 -0.30049 -0.00641 0.39469 1.95947
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.01017 0.05327 -0.191 0.848888
target1 -1.30787 0.15969 -8.190 1.47e-13 ***
target2 -0.55267 0.15969 -3.461 0.000714 ***
target3 0.53573 0.15519 3.452 0.000736 ***
target4 0.55015 0.15519 3.545 0.000534 ***
target5 0.51084 0.16468 3.102 0.002325 **
target6 0.50173 0.15969 3.142 0.002049 **
target7 -0.95033 0.16468 -5.771 4.86e-08 ***
target8 -0.78709 0.15969 -4.929 2.30e-06 ***
target9 0.40228 0.15969 2.519 0.012888 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.6519 on 140 degrees of freedom
Multiple R-squared: 0.6007, Adjusted R-squared: 0.5751
F-statistic: 23.41 on 9 and 140 DF, p-value: < 2.2e-16
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_gh_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-4.3387 -0.3255 0.1168 0.7088 2.5232
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.99365 0.05137 -19.343 < 2e-16 ***
target1 -1.38180 0.19201 -7.196 6.18e-13 ***
target2 -0.59239 0.15996 -3.703 0.000213 ***
target3 0.47928 0.13062 3.669 0.000243 ***
target4 0.55365 0.12986 4.263 2.01e-05 ***
target5 0.53974 0.13765 3.921 8.81e-05 ***
target6 0.54489 0.13357 4.079 4.51e-05 ***
target7 -0.85562 0.17529 -4.881 1.05e-06 ***
target8 -0.59079 0.15990 -3.695 0.000220 ***
target9 0.35280 0.13593 2.595 0.009448 **
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 13.508 1.533 8.81 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 126.7 on 11 Df
Pseudo R-squared: 0.606
Number of iterations: 19 (BFGS) + 2 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + target, data = d_sim_gh_adch)
Residuals:
Min 1Q Median 3Q Max
-2.02493 -0.29739 -0.01966 0.37670 1.84631
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.01504 0.05309 -0.283 0.777448
scale(age) -0.00289 0.05522 -0.052 0.958341
gender_m -0.10258 0.05767 -1.779 0.077482 .
target1 -1.33761 0.16017 -8.351 6.39e-14 ***
target2 -0.54102 0.15906 -3.401 0.000878 ***
target3 0.50267 0.15564 3.230 0.001549 **
target4 0.51660 0.15558 3.320 0.001150 **
target5 0.51554 0.16394 3.145 0.002036 **
target6 0.51338 0.15906 3.228 0.001560 **
target7 -0.93097 0.16425 -5.668 8.11e-08 ***
target8 -0.83032 0.16098 -5.158 8.50e-07 ***
target9 0.45550 0.16237 2.805 0.005754 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.6488 on 138 degrees of freedom
Multiple R-squared: 0.6102, Adjusted R-squared: 0.5791
F-statistic: 19.64 on 11 and 138 DF, p-value: < 2.2e-16Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + +(1 | target)
Data: d_sim_gh_adch
REML criterion at convergence: 333.6
Scaled residuals:
Min 1Q Median 3Q Max
-3.07464 -0.49059 0.01197 0.58928 2.82562
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.6579 0.8111
Residual 0.4209 0.6488
Number of obs: 150, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) -1.441e-02 2.619e-01 8.981e+00 -0.055 0.9573
scale(age) 1.194e-04 5.519e-02 1.382e+02 0.002 0.9983
gender_m -9.769e-02 5.753e-02 1.393e+02 -1.698 0.0917 .
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Correlation of Fixed Effects:
(Intr) scl(g)
scale(age) 0.003
gender_m 0.011 0.252
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + target, data = d_sim_gh_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-4.6476 -0.3971 0.0891 0.7333 2.5570
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.99718 0.05112 -19.505 < 2e-16 ***
scale(age) 0.02744 0.04935 0.556 0.578109
gender_m -0.06856 0.05144 -1.333 0.182585
target1 -1.39723 0.19174 -7.287 3.16e-13 ***
target2 -0.59003 0.15915 -3.707 0.000209 ***
target3 0.45402 0.13075 3.472 0.000516 ***
target4 0.53200 0.12993 4.095 4.23e-05 ***
target5 0.54282 0.13675 3.969 7.21e-05 ***
target6 0.55044 0.13281 4.145 3.40e-05 ***
target7 -0.83935 0.17426 -4.817 1.46e-06 ***
target8 -0.60878 0.16018 -3.801 0.000144 ***
target9 0.37899 0.13827 2.741 0.006126 **
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 13.735 1.559 8.808 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 128 on 13 Df
Pseudo R-squared: 0.6152
Number of iterations: 22 (BFGS) + 1 (Fisher scoring) Thailand
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_th_adch)
Residuals:
Min 1Q Median 3Q Max
-1.5399 -0.7272 -0.2931 0.7147 2.9228
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 4.658e-17 8.100e-02 0.000 1.000
scale(age) -9.654e-02 8.127e-02 -1.188 0.237
Residual standard error: 0.9986 on 150 degrees of freedom
Multiple R-squared: 0.00932, Adjusted R-squared: 0.002716
F-statistic: 1.411 on 1 and 150 DF, p-value: 0.2367
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_th_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.9606 -0.6910 -0.1969 0.7630 2.6049
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.93266 0.04237 -22.01 <2e-16 ***
scale(age) -0.05117 0.04194 -1.22 0.222
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 17.059 1.913 8.916 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 129.1 on 3 Df
Pseudo R-squared: 0.01085
Number of iterations: 9 (BFGS) + 2 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_th_adch)
Residuals:
Min 1Q Median 3Q Max
-1.6021 -0.7579 -0.1796 0.7213 3.1169
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.008783 0.081551 0.108 0.914
gender_m 0.083441 0.081551 1.023 0.308
Residual standard error: 0.9998 on 150 degrees of freedom
Multiple R-squared: 0.006931, Adjusted R-squared: 0.0003104
F-statistic: 1.047 on 1 and 150 DF, p-value: 0.3079
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_th_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.0166 -0.7391 -0.0938 0.7622 2.7135
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.92876 0.04264 -21.782 <2e-16 ***
gender_m 0.03456 0.04205 0.822 0.411
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 16.964 1.902 8.918 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 128.7 on 3 Df
Pseudo R-squared: 0.004708
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) Religion
Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_th_adch, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-1.5939 -0.6810 -0.2017 0.4187 2.8156
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.001869 0.078903 -0.024 0.9811
target1 -0.022011 0.238036 -0.092 0.9265
target2 -0.231547 0.231320 -1.001 0.3185
target3 -0.019541 0.238036 -0.082 0.9347
target4 -0.276511 0.238036 -1.162 0.2473
target5 -0.352161 0.238036 -1.479 0.1412
target6 -0.109907 0.238036 -0.462 0.6450
target7 -0.399233 0.238036 -1.677 0.0957 .
target8 0.452051 0.238036 1.899 0.0596 .
target9 0.515583 0.231320 2.229 0.0274 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9725 on 142 degrees of freedom
Multiple R-squared: 0.1107, Adjusted R-squared: 0.05433
F-statistic: 1.964 on 9 and 142 DF, p-value: 0.04774
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_th_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.8769 -0.7085 -0.1184 0.5032 2.7180
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.939353 0.040478 -23.206 <2e-16 ***
target1 0.001302 0.120112 0.011 0.9914
target2 -0.096821 0.118742 -0.815 0.4148
target3 -0.016916 0.120483 -0.140 0.8883
target4 -0.163343 0.123739 -1.320 0.1868
target5 -0.168808 0.123870 -1.363 0.1729
target6 -0.048193 0.121139 -0.398 0.6908
target7 -0.232307 0.125437 -1.852 0.0640 .
target8 0.231012 0.116053 1.991 0.0465 *
target9 0.237074 0.112724 2.103 0.0355 *
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 19.003 2.136 8.895 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 137.2 on 11 Df
Pseudo R-squared: 0.116
Number of iterations: 18 (BFGS) + 2 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + target, data = d_sim_th_adch)
Residuals:
Min 1Q Median 3Q Max
-1.7560 -0.6538 -0.2126 0.5222 2.9321
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.005421 0.079310 0.068 0.9456
scale(age) -0.103910 0.079957 -1.300 0.1959
gender_m 0.068863 0.080833 0.852 0.3957
target1 -0.047832 0.239150 -0.200 0.8418
target2 -0.237626 0.231949 -1.024 0.3074
target3 -0.010505 0.238525 -0.044 0.9649
target4 -0.283076 0.238106 -1.189 0.2365
target5 -0.333051 0.238144 -1.399 0.1642
target6 -0.094453 0.238499 -0.396 0.6927
target7 -0.389305 0.238023 -1.636 0.1042
target8 0.458324 0.237920 1.926 0.0561 .
target9 0.515521 0.232123 2.221 0.0280 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9715 on 140 degrees of freedom
Multiple R-squared: 0.1249, Adjusted R-squared: 0.05617
F-statistic: 1.817 on 11 and 140 DF, p-value: 0.05638Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + +(1 | target)
Data: d_sim_th_adch
REML criterion at convergence: 435.1
Scaled residuals:
Min 1Q Median 3Q Max
-1.5498 -0.7577 -0.2870 0.6243 3.0218
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.05678 0.2383
Residual 0.94357 0.9714
Number of obs: 152, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 0.007629 0.109367 9.128123 0.070 0.946
scale(age) -0.103718 0.079594 142.376071 -1.303 0.195
gender_m 0.080764 0.080103 144.424864 1.008 0.315
Correlation of Fixed Effects:
(Intr) scl(g)
scale(age) -0.005
gender_m 0.078 -0.072
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + target, data = d_sim_th_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.1108 -0.7106 -0.1652 0.6362 2.8274
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.93763 0.04040 -23.206 <2e-16 ***
scale(age) -0.05734 0.04014 -1.428 0.1532
gender_m 0.02530 0.04055 0.624 0.5327
target1 -0.01090 0.11989 -0.091 0.9276
target2 -0.09795 0.11838 -0.827 0.4080
target3 -0.01588 0.12007 -0.132 0.8948
target4 -0.16828 0.12307 -1.367 0.1715
target5 -0.15906 0.12317 -1.291 0.1966
target6 -0.04316 0.12067 -0.358 0.7206
target7 -0.23011 0.12479 -1.844 0.0652 .
target8 0.23550 0.11528 2.043 0.0411 *
target9 0.24164 0.11245 2.149 0.0316 *
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 19.303 2.171 8.892 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 138.4 on 13 Df
Pseudo R-squared: 0.1297
Number of iterations: 22 (BFGS) + 1 (Fisher scoring) China
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_ch_adch)
Residuals:
Min 1Q Median 3Q Max
-1.7903 -0.7533 -0.1242 0.7703 2.4331
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.000761 0.087667 0.009 0.993
scale(age) 0.093244 0.088050 1.059 0.292
Residual standard error: 0.9995 on 128 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.008685, Adjusted R-squared: 0.0009408
F-statistic: 1.121 on 1 and 128 DF, p-value: 0.2916
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_ch_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.8868 -0.6674 0.0562 0.8073 1.9696
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.09024 0.05771 -18.890 <2e-16 ***
scale(age) 0.05319 0.05616 0.947 0.344
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 11.214 1.352 8.294 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 93.52 on 3 Df
Pseudo R-squared: 0.007469
Number of iterations: 14 (BFGS) + 1 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_ch_adch)
Residuals:
Min 1Q Median 3Q Max
-1.6607 -0.8233 -0.1631 0.8235 2.3939
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.0004534 0.0880196 -0.005 0.996
gender_m -0.0294729 0.0880196 -0.335 0.738
Residual standard error: 1.003 on 128 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.0008752, Adjusted R-squared: -0.00693
F-statistic: 0.1121 on 1 and 128 DF, p-value: 0.7383
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_ch_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.6994 -0.7133 0.0207 0.8461 1.9501
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.09034 0.05788 -18.837 <2e-16 ***
gender_m -0.01502 0.05613 -0.268 0.789
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 11.139 1.343 8.296 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 93.1 on 3 Df
Pseudo R-squared: 0.000565
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_ch_adch, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-1.84233 -0.58785 -0.08649 0.58315 2.57595
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.04026 0.11436 0.352 0.72541
target1 0.01966 0.26276 0.075 0.94050
target2 0.20515 0.25504 0.804 0.42278
target3 0.47792 0.86061 0.555 0.57972
target4 -0.21330 0.28146 -0.758 0.45005
target5 -0.11062 0.25504 -0.434 0.66528
target6 0.33042 0.26276 1.257 0.21105
target7 -0.51317 0.27149 -1.890 0.06117 .
target8 -0.82181 0.26276 -3.128 0.00222 **
target9 0.20730 0.25504 0.813 0.41795
target10 -0.32229 0.26276 -1.227 0.22242
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.943 on 119 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.1797, Adjusted R-squared: 0.1108
F-statistic: 2.607 on 10 and 119 DF, p-value: 0.006741All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + target, data = d_sim_ch_adch)
Residuals:
Min 1Q Median 3Q Max
-1.7216 -0.5766 -0.0338 0.5585 2.5411
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.02634 0.11524 0.229 0.81961
scale(age) 0.07098 0.08755 0.811 0.41914
gender_m -0.08365 0.08618 -0.971 0.33370
target1 0.05474 0.26524 0.206 0.83685
target2 0.18557 0.25797 0.719 0.47336
target3 0.33288 0.87154 0.382 0.70320
target4 -0.15307 0.28679 -0.534 0.59454
target5 -0.06450 0.25869 -0.249 0.80355
target6 0.33335 0.26414 1.262 0.20946
target7 -0.53974 0.27314 -1.976 0.05050 .
target8 -0.82533 0.26377 -3.129 0.00221 **
target9 0.25342 0.25886 0.979 0.32961
target10 -0.31938 0.26620 -1.200 0.23264
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9453 on 117 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.1895, Adjusted R-squared: 0.1063
F-statistic: 2.279 on 12 and 117 DF, p-value: 0.01223Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + +(1 | target)
Data: d_sim_ch_adch
REML criterion at convergence: 369.6
Scaled residuals:
Min 1Q Median 3Q Max
-1.6602 -0.7118 -0.1095 0.6741 2.5098
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.1285 0.3585
Residual 0.8890 0.9429
Number of obs: 130, groups: target, 11
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 0.001534 0.139281 9.354759 0.011 0.991
scale(age) 0.082578 0.085695 123.487072 0.964 0.337
gender_m -0.069777 0.084725 122.182416 -0.824 0.412
Correlation of Fixed Effects:
(Intr) scl(g)
scale(age) -0.004
gender_m 0.017 -0.122Vanuatu
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_vt_adch)
Residuals:
Min 1Q Median 3Q Max
-1.8608 -0.9267 0.1151 0.8844 1.4462
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.01157 0.08368 0.138 0.890
scale(age) 0.02228 0.08398 0.265 0.791
Residual standard error: 0.9972 on 140 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.0005025, Adjusted R-squared: -0.006637
F-statistic: 0.07038 on 1 and 140 DF, p-value: 0.7912
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_vt_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.1805 -0.8564 0.1844 0.8695 1.3754
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.430248 0.048815 -8.814 <2e-16 ***
scale(age) 0.006664 0.048773 0.137 0.891
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 11.32 1.29 8.771 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 80.22 on 3 Df
Pseudo R-squared: 0.0001346
Number of iterations: 9 (BFGS) + 2 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_vt_adch)
Residuals:
Min 1Q Median 3Q Max
-1.86647 -0.90478 0.09592 0.88104 1.46081
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.008161 0.084711 -0.096 0.923
gender_m -0.055574 0.084711 -0.656 0.513
Residual standard error: 1.002 on 141 degrees of freedom
Multiple R-squared: 0.003043, Adjusted R-squared: -0.004028
F-statistic: 0.4304 on 1 and 141 DF, p-value: 0.5129
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_vt_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.1830 -0.8390 0.1678 0.8723 1.3920
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.44208 0.04950 -8.931 <2e-16 ***
gender_m -0.03318 0.04928 -0.673 0.501
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 11.198 1.272 8.805 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 80.26 on 3 Df
Pseudo R-squared: 0.003177
Number of iterations: 11 (BFGS) + 2 (Fisher scoring) Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_vt_adch, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-2.22143 -0.59576 0.05297 0.63392 2.01145
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.02832 0.07872 0.360 0.71964
target1 -0.67395 0.23952 -2.814 0.00565 **
target2 -0.51665 0.23176 -2.229 0.02749 *
target3 0.29425 0.29892 0.984 0.32672
target4 0.38686 0.34215 1.131 0.26025
target5 0.23323 0.22483 1.037 0.30145
target6 0.36705 0.23176 1.584 0.11565
target7 0.55082 0.21857 2.520 0.01293 *
target8 -0.95911 0.22483 -4.266 3.77e-05 ***
target9 -0.13908 0.22483 -0.619 0.53723
target10 0.09963 0.23952 0.416 0.67810
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9017 on 132 degrees of freedom
Multiple R-squared: 0.2442, Adjusted R-squared: 0.1869
F-statistic: 4.265 on 10 and 132 DF, p-value: 3.732e-05
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_vt_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.0861 -0.6609 0.1370 0.7399 2.2219
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.42754 0.04514 -9.472 < 2e-16 ***
target1 -0.42361 0.14315 -2.959 0.00309 **
target2 -0.31615 0.13636 -2.318 0.02043 *
target3 0.19121 0.16786 1.139 0.25467
target4 0.24399 0.19166 1.273 0.20300
target5 0.13825 0.12671 1.091 0.27527
target6 0.21419 0.13015 1.646 0.09981 .
target7 0.31819 0.12241 2.599 0.00934 **
target8 -0.56957 0.13749 -4.143 3.43e-05 ***
target9 -0.06163 0.12854 -0.479 0.63159
target10 0.05224 0.13569 0.385 0.70023
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 14.787 1.696 8.716 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 100 on 12 Df
Pseudo R-squared: 0.2479
Number of iterations: 21 (BFGS) + 2 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + target, data = d_sim_vt_adch)
Residuals:
Min 1Q Median 3Q Max
-2.23284 -0.57970 0.03138 0.58494 1.98058
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.03605 0.07853 0.459 0.64698
scale(age) -0.00728 0.07643 -0.095 0.92427
gender_m -0.04021 0.07775 -0.517 0.60588
target1 -0.68118 0.23828 -2.859 0.00496 **
target2 -0.51908 0.22956 -2.261 0.02542 *
target3 0.27485 0.29517 0.931 0.35352
target4 0.38024 0.33757 1.126 0.26209
target5 0.22451 0.22239 1.010 0.31461
target6 0.34737 0.22874 1.519 0.13130
target7 0.67889 0.22244 3.052 0.00276 **
target8 -0.98538 0.22306 -4.418 2.09e-05 ***
target9 -0.16564 0.22303 -0.743 0.45903
target10 0.10364 0.23916 0.433 0.66548
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8885 on 129 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.2688, Adjusted R-squared: 0.2008
F-statistic: 3.953 on 12 and 129 DF, p-value: 3.458e-05Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + +(1 | target)
Data: d_sim_vt_adch
REML criterion at convergence: 391.1
Scaled residuals:
Min 1Q Median 3Q Max
-2.44409 -0.81803 0.08568 0.71906 2.05346
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.2182 0.4672
Residual 0.7880 0.8877
Number of obs: 142, groups: target, 11
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 0.025394 0.160746 10.417215 0.158 0.877
scale(age) -0.000469 0.076012 131.630258 -0.006 0.995
gender_m -0.041024 0.077200 132.360377 -0.531 0.596
Correlation of Fixed Effects:
(Intr) scl(g)
scale(age) -0.001
gender_m 0.070 -0.035
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + target, data = d_sim_vt_adch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.2043 -0.6984 0.0942 0.7788 2.3123
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.42257 0.04466 -9.463 < 2e-16 ***
scale(age) -0.00811 0.04345 -0.187 0.85194
gender_m -0.01615 0.04421 -0.365 0.71491
target1 -0.43278 0.14123 -3.064 0.00218 **
target2 -0.32141 0.13397 -2.399 0.01644 *
target3 0.17955 0.16438 1.092 0.27470
target4 0.23957 0.18750 1.278 0.20134
target5 0.13341 0.12430 1.073 0.28316
target6 0.20318 0.12737 1.595 0.11068
target7 0.40402 0.12340 3.274 0.00106 **
target8 -0.58371 0.13530 -4.314 1.6e-05 ***
target9 -0.07637 0.12648 -0.604 0.54598
target10 0.05178 0.13439 0.385 0.70005
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 15.545 1.792 8.674 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 102.6 on 14 Df
Pseudo R-squared: 0.2747
Number of iterations: 22 (BFGS) + 2 (Fisher scoring) Using US model for other adults (kinda = yes = 1)
Unequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vector
Unequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorUnequal factor levels: coercing to characterbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vectorbinding character and factor vector, coercing into character vector
Ghana
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_us_gh)
Residuals:
Min 1Q Median 3Q Max
-1.2684 -0.9444 -0.1291 1.0319 1.8814
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.02676 0.08152 -0.328 0.7432
scale(age) -0.18392 0.08180 -2.248 0.0261 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.985 on 144 degrees of freedom
(4 observations deleted due to missingness)
Multiple R-squared: 0.03392, Adjusted R-squared: 0.02721
F-statistic: 5.055 on 1 and 144 DF, p-value: 0.02607
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_us_gh)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.3607 -0.9458 0.1580 0.9800 1.4586
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.05011 0.07873 -13.339 <2e-16 ***
scale(age) -0.14764 0.07474 -1.975 0.0482 *
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 4.6320 0.5147 9 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 63.1 on 3 Df
Pseudo R-squared: 0.02768
Number of iterations: 12 (BFGS) + 1 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_us_gh)
Residuals:
Min 1Q Median 3Q Max
-1.14287 -0.86431 -0.03598 1.04135 1.66985
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.01671 0.08172 -0.205 0.8382
gender_m -0.13928 0.08172 -1.704 0.0904 .
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9937 on 148 degrees of freedom
Multiple R-squared: 0.01925, Adjusted R-squared: 0.01262
F-statistic: 2.904 on 1 and 148 DF, p-value: 0.09043
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_us_gh)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.1781 -0.8583 0.2286 0.9768 1.3478
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.03667 0.07848 -13.209 <2e-16 ***
gender_m -0.10650 0.07351 -1.449 0.147
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 4.5516 0.4974 9.152 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 62.04 on 3 Df
Pseudo R-squared: 0.01585
Number of iterations: 13 (BFGS) + 1 (Fisher scoring) Race/ethnicity
Call:
lm(formula = scale(MSE) ~ ethnicity_cat2, data = d_sim_us_gh)
Residuals:
Min 1Q Median 3Q Max
-1.0738 -0.9378 -0.1646 1.0779 1.4607
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.01451 0.08367 -0.173 0.863
ethnicity_cat2_nonFante -0.06803 0.08367 -0.813 0.417
Residual standard error: 1.001 on 148 degrees of freedom
Multiple R-squared: 0.004447, Adjusted R-squared: -0.00228
F-statistic: 0.6611 on 1 and 148 DF, p-value: 0.4175
Call:
betareg(formula = MSE_rescaled ~ ethnicity_cat2, data = d_sim_us_gh)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.0986 -0.9205 0.1272 0.9904 1.2336
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.03527 0.07985 -12.966 <2e-16 ***
ethnicity_cat2_nonFante -0.06020 0.07486 -0.804 0.421
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 4.506 0.492 9.159 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 61.31 on 3 Df
Pseudo R-squared: 0.00493
Number of iterations: 12 (BFGS) + 1 (Fisher scoring) Education
Call:
lm(formula = scale(MSE) ~ scale(education_catX), data = d_sim_us_gh %>%
mutate(education_catX = as.numeric(education_catX)))
Residuals:
Min 1Q Median 3Q Max
-1.0867 -1.0339 -0.2125 1.1076 1.5410
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.006848 0.081976 0.084 0.934
scale(education_catX) -0.071715 0.082253 -0.872 0.385
Residual standard error: 1.001 on 147 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.005145, Adjusted R-squared: -0.001623
F-statistic: 0.7602 on 1 and 147 DF, p-value: 0.3847
Call:
betareg(formula = MSE_rescaled ~ scale(education_catX), data = d_sim_us_gh %>% mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.1264 -1.0452 0.0822 1.0219 1.2991
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.01661 0.07802 -13.029 <2e-16 ***
scale(education_catX) -0.07037 0.07395 -0.952 0.341
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 4.5143 0.4943 9.132 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 60.58 on 3 Df
Pseudo R-squared: 0.006762
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) 
Call:
lm(formula = scale(MSE) ~ education_cat2, data = d_sim_us_gh)
Residuals:
Min 1Q Median 3Q Max
-1.1620 -0.9139 -0.1157 1.0655 1.5321
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.01767 0.08187 0.216 0.829
education_cat2_hs -0.12403 0.08187 -1.515 0.132
Residual standard error: 0.9955 on 147 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.01537, Adjusted R-squared: 0.008677
F-statistic: 2.295 on 1 and 147 DF, p-value: 0.1319
Call:
betareg(formula = MSE_rescaled ~ education_cat2, data = d_sim_us_gh)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.2308 -0.8942 0.1770 0.9836 1.2787
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.00834 0.07782 -12.957 <2e-16 ***
education_cat2_hs -0.11043 0.07311 -1.511 0.131
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 4.5572 0.4994 9.125 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 61.26 on 3 Df
Pseudo R-squared: 0.01691
Number of iterations: 13 (BFGS) + 1 (Fisher scoring) Rural/urban
Call:
lm(formula = scale(MSE) ~ urban_rural_cat2, data = d_sim_us_gh)
Residuals:
Min 1Q Median 3Q Max
-1.2025 -0.9117 -0.2291 1.0373 1.6224
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.03683 0.08384 0.439 0.661
urban_rural_cat2_rural 0.14538 0.08384 1.734 0.085 .
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9933 on 148 degrees of freedom
Multiple R-squared: 0.01991, Adjusted R-squared: 0.01329
F-statistic: 3.007 on 1 and 148 DF, p-value: 0.085
Call:
betareg(formula = MSE_rescaled ~ urban_rural_cat2, data = d_sim_us_gh)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.3064 -0.8883 0.0595 0.9674 1.3400
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.99079 0.07895 -12.549 <2e-16 ***
urban_rural_cat2_rural 0.13563 0.07460 1.818 0.069 .
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 4.5878 0.5016 9.146 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 62.6 on 3 Df
Pseudo R-squared: 0.02371
Number of iterations: 13 (BFGS) + 1 (Fisher scoring) Religion
Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_us_gh, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-1.38302 -0.18471 0.00000 0.09013 1.88009
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -2.448e-16 4.120e-02 0.000 1.00000
target1 -1.020e+00 1.236e-01 -8.254 1.02e-13 ***
target2 -1.020e+00 1.236e-01 -8.254 1.02e-13 ***
target3 -1.565e-01 1.236e-01 -1.266 0.20764
target4 3.627e-01 1.236e-01 2.934 0.00391 **
target5 9.190e-01 1.236e-01 7.435 9.50e-12 ***
target6 1.256e+00 1.236e-01 10.165 < 2e-16 ***
target7 -1.050e-01 1.236e-01 -0.849 0.39726
target8 -1.008e+00 1.236e-01 -8.157 1.77e-13 ***
target9 -4.999e-01 1.236e-01 -4.044 8.64e-05 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.5046 on 140 degrees of freedom
Multiple R-squared: 0.7607, Adjusted R-squared: 0.7454
F-statistic: 49.46 on 9 and 140 DF, p-value: < 2.2e-16
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_us_gh)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.8507 -0.2602 0.0000 0.2276 3.0270
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.20755 0.04864 -24.829 < 2e-16 ***
target1 -1.12750 0.17136 -6.580 4.72e-11 ***
target2 -1.12750 0.17136 -6.580 4.72e-11 ***
target3 -0.04954 0.13161 -0.376 0.707
target4 0.51130 0.12017 4.255 2.09e-05 ***
target5 1.01276 0.11573 8.751 < 2e-16 ***
target6 1.26792 0.11554 10.974 < 2e-16 ***
target7 -0.08394 0.13254 -0.633 0.526
target8 -1.10397 0.17031 -6.482 9.04e-11 ***
target9 -0.57896 0.14857 -3.897 9.74e-05 ***
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 18.419 2.115 8.707 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 162.1 on 11 Df
Pseudo R-squared: 0.7615
Number of iterations: 20 (BFGS) + 1 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + scale(education_catX) +
ethnicity_cat2 + urban_rural_cat2 + target, data = d_sim_us_gh %>%
mutate(education_catX = as.numeric(education_catX)))
Residuals:
Min 1Q Median 3Q Max
-1.24912 -0.15051 -0.01104 0.12924 1.83348
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.002067 0.045563 0.045 0.963892
scale(age) -0.051117 0.047703 -1.072 0.285908
gender_m 0.018945 0.045660 0.415 0.678881
scale(education_catX) 0.027478 0.050714 0.542 0.588864
ethnicity_cat2_nonFante -0.058514 0.047792 -1.224 0.223035
urban_rural_cat2_rural 0.058046 0.055038 1.055 0.293544
target1 -0.953915 0.132597 -7.194 4.45e-11 ***
target2 -1.040403 0.128308 -8.109 3.32e-13 ***
target3 -0.146720 0.129901 -1.129 0.260777
target4 0.332184 0.130571 2.544 0.012127 *
target5 0.925414 0.133589 6.927 1.78e-10 ***
target6 1.225074 0.136186 8.996 2.43e-15 ***
target7 -0.092543 0.128999 -0.717 0.474418
target8 -1.007726 0.132126 -7.627 4.49e-12 ***
target9 -0.502963 0.130329 -3.859 0.000178 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.5138 on 130 degrees of freedom
(5 observations deleted due to missingness)
Multiple R-squared: 0.7611, Adjusted R-squared: 0.7354
F-statistic: 29.58 on 14 and 130 DF, p-value: < 2.2e-16
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + education_cat2 +
ethnicity_cat2 + urban_rural_cat2 + target, data = d_sim_us_gh)
Residuals:
Min 1Q Median 3Q Max
-1.25597 -0.13694 -0.00908 0.12961 1.84592
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.000315 0.045445 0.007 0.994480
scale(age) -0.046497 0.047768 -0.973 0.332165
gender_m 0.019951 0.045844 0.435 0.664138
education_cat2_hs 0.019182 0.053122 0.361 0.718617
ethnicity_cat2_nonFante -0.054516 0.046884 -1.163 0.247051
urban_rural_cat2_rural 0.053244 0.055421 0.961 0.338481
target1 -0.957406 0.132451 -7.228 3.72e-11 ***
target2 -1.037976 0.128361 -8.086 3.74e-13 ***
target3 -0.147210 0.130159 -1.131 0.260137
target4 0.333238 0.130849 2.547 0.012039 *
target5 0.928684 0.133442 6.959 1.51e-10 ***
target6 1.231436 0.137476 8.957 3.01e-15 ***
target7 -0.096081 0.131384 -0.731 0.465914
target8 -1.009013 0.132177 -7.634 4.33e-12 ***
target9 -0.509947 0.129675 -3.933 0.000136 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.5141 on 130 degrees of freedom
(5 observations deleted due to missingness)
Multiple R-squared: 0.7608, Adjusted R-squared: 0.735
F-statistic: 29.53 on 14 and 130 DF, p-value: < 2.2e-16Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + scale(education_catX) + ethnicity_cat2 +
urban_rural_cat2 + +(1 | target)
Data: d_sim_us_gh %>% mutate(education_catX = as.numeric(education_catX))
REML criterion at convergence: 272.4
Scaled residuals:
Min 1Q Median 3Q Max
-2.4160 -0.2900 -0.0277 0.2694 3.5324
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.7958 0.8921
Residual 0.2640 0.5138
Number of obs: 145, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 0.001911 0.285760 9.017062 0.007 0.995
scale(age) -0.055075 0.047641 130.669246 -1.156 0.250
gender_m 0.016428 0.045624 130.408826 0.360 0.719
scale(education_catX) 0.028365 0.050690 130.235946 0.560 0.577
ethnicity_cat2_nonFante -0.060290 0.047729 130.680068 -1.263 0.209
urban_rural_cat2_rural 0.062004 0.054983 130.512288 1.128 0.262
Correlation of Fixed Effects:
(Intr) scl(g) gndr_m sc(_X) et_2_F
scale(age) -0.009
gender_m 0.018 -0.220
scl(dctn_X) 0.013 -0.104 0.015
ethncty_2_F 0.025 0.106 -0.047 -0.221
urbn_rrl_2_ 0.040 -0.250 0.074 0.508 -0.220Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + education_cat2 + ethnicity_cat2 +
urban_rural_cat2 + +(1 | target)
Data: d_sim_us_gh
REML criterion at convergence: 272.6
Scaled residuals:
Min 1Q Median 3Q Max
-2.4302 -0.2740 -0.0384 0.2519 3.5530
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.8002 0.8945
Residual 0.2643 0.5141
Number of obs: 145, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 6.155e-05 2.865e-01 9.011e+00 0.000 1.000
scale(age) -5.060e-02 4.770e-02 1.307e+02 -1.061 0.291
gender_m 1.726e-02 4.581e-02 1.304e+02 0.377 0.707
education_cat2_hs 1.688e-02 5.307e-02 1.305e+02 0.318 0.751
ethnicity_cat2_nonFante -5.590e-02 4.682e-02 1.307e+02 -1.194 0.235
urban_rural_cat2_rural 5.547e-02 5.537e-02 1.305e+02 1.002 0.318
Correlation of Fixed Effects:
(Intr) scl(g) gndr_m edc_2_ et_2_F
scale(age) -0.008
gender_m 0.018 -0.208
edctn_ct2_h 0.003 0.110 0.083
ethncty_2_F 0.029 0.074 -0.053 -0.102
urbn_rrl_2_ 0.034 -0.139 0.108 0.517 -0.161
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + scale(education_catX) + ethnicity_cat2 +
urban_rural_cat2 + target, data = d_sim_us_gh %>% mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.4591 -0.2235 -0.0073 0.2487 3.0691
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.20724 0.05190 -23.260 < 2e-16 ***
scale(age) -0.04927 0.04894 -1.007 0.314064
gender_m 0.03805 0.04819 0.790 0.429753
scale(education_catX) 0.03536 0.05442 0.650 0.515900
ethnicity_cat2_nonFante -0.07092 0.05079 -1.396 0.162588
urban_rural_cat2_rural 0.05223 0.05761 0.907 0.364650
target1 -1.06368 0.17637 -6.031 1.63e-09 ***
target2 -1.16250 0.17355 -6.698 2.11e-11 ***
target3 -0.01660 0.13537 -0.123 0.902419
target4 0.48504 0.12517 3.875 0.000107 ***
target5 1.01682 0.12311 8.260 < 2e-16 ***
target6 1.23652 0.12461 9.923 < 2e-16 ***
target7 -0.07233 0.13549 -0.534 0.593427
target8 -1.11796 0.17766 -6.293 3.12e-10 ***
target9 -0.58252 0.15200 -3.832 0.000127 ***
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 18.475 2.159 8.557 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 157.8 on 16 Df
Pseudo R-squared: 0.7615
Number of iterations: 24 (BFGS) + 2 (Fisher scoring)
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + education_cat2 + ethnicity_cat2 +
urban_rural_cat2 + target, data = d_sim_us_gh %>% mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.4775 -0.2591 0.0059 0.2671 3.0730
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.20900 0.05186 -23.315 < 2e-16 ***
scale(age) -0.04357 0.04889 -0.891 0.372797
gender_m 0.03892 0.04841 0.804 0.421395
education_cat2_hs 0.02452 0.05498 0.446 0.655619
ethnicity_cat2_nonFante -0.06402 0.04916 -1.302 0.192781
urban_rural_cat2_rural 0.04500 0.05695 0.790 0.429383
target1 -1.06732 0.17629 -6.054 1.41e-09 ***
target2 -1.15787 0.17348 -6.675 2.48e-11 ***
target3 -0.01637 0.13569 -0.121 0.903967
target4 0.48449 0.12549 3.861 0.000113 ***
target5 1.02059 0.12295 8.301 < 2e-16 ***
target6 1.24470 0.12612 9.869 < 2e-16 ***
target7 -0.07905 0.13794 -0.573 0.566600
target8 -1.11841 0.17770 -6.294 3.10e-10 ***
target9 -0.59084 0.15152 -3.900 9.64e-05 ***
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 18.438 2.155 8.557 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 157.7 on 16 Df
Pseudo R-squared: 0.7614
Number of iterations: 24 (BFGS) + 2 (Fisher scoring) Thailand
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_us_th)
Residuals:
Min 1Q Median 3Q Max
-1.8652 -0.8126 0.1061 0.9638 1.4156
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.002823 0.082393 0.034 0.973
scale(age) 0.029688 0.082671 0.359 0.720
Residual standard error: 1.006 on 147 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.0008765, Adjusted R-squared: -0.00592
F-statistic: 0.129 on 1 and 147 DF, p-value: 0.72
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_us_th)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.3632 -0.5967 0.2285 0.9066 1.2356
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.667272 0.057801 -11.544 <2e-16 ***
scale(age) 0.009412 0.057056 0.165 0.869
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 7.8806 0.8674 9.085 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 68.03 on 3 Df
Pseudo R-squared: 0.0001609
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_us_th)
Residuals:
Min 1Q Median 3Q Max
-1.87720 -0.68196 0.04427 0.92086 1.39444
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.4503 0.2404 -1.873 0.0630 .
gender_m 0.4007 0.2515 1.593 0.1132
gender_o -0.9159 0.4710 -1.944 0.0538 .
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9923 on 147 degrees of freedom
Multiple R-squared: 0.02853, Adjusted R-squared: 0.01531
F-statistic: 2.159 on 2 and 147 DF, p-value: 0.1191
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_us_th)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.4240 -0.5098 0.1910 0.8823 1.2392
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.9693 0.1918 -5.054 4.32e-07 ***
gender_m 0.2776 0.1979 1.402 0.161
gender_o -0.6122 0.3766 -1.625 0.104
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 8.1244 0.8928 9.1 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 70.72 on 4 Df
Pseudo R-squared: 0.02693
Number of iterations: 15 (BFGS) + 1 (Fisher scoring) Race/ethnicity
Education
Call:
lm(formula = scale(MSE) ~ scale(education_catX), data = d_sim_us_th %>%
mutate(education_catX = as.numeric(education_catX)))
Residuals:
Min 1Q Median 3Q Max
-1.9723 -0.8418 0.1319 0.9013 1.5219
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.01392 0.08298 -0.168 0.867
scale(education_catX) -0.13110 0.08327 -1.574 0.118
Residual standard error: 0.9992 on 143 degrees of freedom
(5 observations deleted due to missingness)
Multiple R-squared: 0.01704, Adjusted R-squared: 0.01016
F-statistic: 2.479 on 1 and 143 DF, p-value: 0.1176
Call:
betareg(formula = MSE_rescaled ~ scale(education_catX), data = d_sim_us_th %>% mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.5294 -0.6501 0.2511 0.8554 1.3312
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.67935 0.05850 -11.612 <2e-16 ***
scale(education_catX) -0.08343 0.05760 -1.448 0.148
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 7.9553 0.8884 8.955 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 67.34 on 3 Df
Pseudo R-squared: 0.01378
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) 
Call:
lm(formula = scale(MSE) ~ education_cat2, data = d_sim_us_th)
Residuals:
Min 1Q Median 3Q Max
-1.9310 -0.8785 0.1087 0.9373 1.4996
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.005588 0.084440 0.066 0.947
education_cat2_coll -0.113134 0.084440 -1.340 0.182
Residual standard error: 1.002 on 143 degrees of freedom
(5 observations deleted due to missingness)
Multiple R-squared: 0.0124, Adjusted R-squared: 0.005491
F-statistic: 1.795 on 1 and 143 DF, p-value: 0.1824
Call:
betareg(formula = MSE_rescaled ~ education_cat2, data = d_sim_us_th)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.4762 -0.6956 0.2506 0.8834 1.3147
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.66633 0.05924 -11.248 <2e-16 ***
education_cat2_coll -0.07441 0.05837 -1.275 0.202
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 7.9288 0.8852 8.957 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 67.09 on 3 Df
Pseudo R-squared: 0.01053
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) Rural/urban
Call:
lm(formula = scale(MSE) ~ urban_rural_cat2, data = d_sim_us_th)
Residuals:
Min 1Q Median 3Q Max
-1.8234 -0.7815 0.1009 0.9694 1.3836
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.01021 0.08940 -0.114 0.909
urban_rural_cat2_rural 0.02133 0.08940 0.239 0.812
Residual standard error: 0.9954 on 139 degrees of freedom
(9 observations deleted due to missingness)
Multiple R-squared: 0.0004094, Adjusted R-squared: -0.006782
F-statistic: 0.05693 on 1 and 139 DF, p-value: 0.8118
Call:
betareg(formula = MSE_rescaled ~ urban_rural_cat2, data = d_sim_us_th)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.3571 -0.5830 0.2253 0.9259 1.2419
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.6702905 0.0625967 -10.708 <2e-16 ***
urban_rural_cat2_rural 0.0008962 0.0617602 0.015 0.988
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 8.0807 0.9154 8.828 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 65.85 on 3 Df
Pseudo R-squared: 1.428e-06
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) Religion
Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_us_th, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-2.1383 -0.6608 0.0856 0.7984 1.7641
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 3.626e-17 7.663e-02 0.000 1.0000
target1 3.058e-01 2.299e-01 1.330 0.1856
target2 2.844e-01 2.299e-01 1.237 0.2181
target3 4.476e-01 2.299e-01 1.947 0.0535 .
target4 -1.659e-01 2.299e-01 -0.722 0.4718
target5 7.442e-02 2.299e-01 0.324 0.7467
target6 -2.284e-01 2.299e-01 -0.993 0.3223
target7 -1.065e+00 2.299e-01 -4.634 8.12e-06 ***
target8 3.259e-01 2.299e-01 1.418 0.1585
target9 8.166e-02 2.299e-01 0.355 0.7230
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9386 on 140 degrees of freedom
Multiple R-squared: 0.1723, Adjusted R-squared: 0.1191
F-statistic: 3.238 on 9 and 140 DF, p-value: 0.001335
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_us_th)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.1430 -0.5794 0.1988 0.8286 1.6591
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.67999 0.05348 -12.715 < 2e-16 ***
target1 0.19111 0.15426 1.239 0.2154
target2 0.16601 0.15458 1.074 0.2829
target3 0.33224 0.15279 2.174 0.0297 *
target4 -0.08796 0.15868 -0.554 0.5794
target5 0.09443 0.15557 0.607 0.5439
target6 -0.10005 0.15892 -0.630 0.5290
target7 -0.78145 0.17758 -4.401 1.08e-05 ***
target8 0.22629 0.15385 1.471 0.1413
target9 0.01213 0.15687 0.077 0.9383
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 9.561 1.059 9.026 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 82.79 on 11 Df
Pseudo R-squared: 0.1741
Number of iterations: 21 (BFGS) + 2 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + scale(education_catX) +
urban_rural_cat2 + target, data = d_sim_us_th %>% mutate(education_catX = as.numeric(education_catX)))
Residuals:
Min 1Q Median 3Q Max
-2.38074 -0.61920 0.04159 0.68964 1.78708
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.412495 0.234763 -1.757 0.0814 .
scale(age) -0.050702 0.098004 -0.517 0.6059
gender_m 0.372744 0.243310 1.532 0.1281
gender_o -0.817959 0.458864 -1.783 0.0772 .
scale(education_catX) -0.166997 0.099411 -1.680 0.0956 .
urban_rural_cat2_rural -0.019313 0.089466 -0.216 0.8294
target1 0.386281 0.243392 1.587 0.1151
target2 0.155380 0.258712 0.601 0.5492
target3 0.468897 0.234203 2.002 0.0475 *
target4 -0.005458 0.241182 -0.023 0.9820
target5 0.046755 0.248550 0.188 0.8511
target6 -0.128337 0.236007 -0.544 0.5876
target7 -0.983734 0.238600 -4.123 6.89e-05 ***
target8 0.368886 0.240734 1.532 0.1280
target9 -0.130218 0.274256 -0.475 0.6358
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9327 on 121 degrees of freedom
(14 observations deleted due to missingness)
Multiple R-squared: 0.2146, Adjusted R-squared: 0.1237
F-statistic: 2.361 on 14 and 121 DF, p-value: 0.006243
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + education_cat2 +
urban_rural_cat2 + target, data = d_sim_us_th)
Residuals:
Min 1Q Median 3Q Max
-2.35168 -0.64247 0.05175 0.69392 1.75683
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.39817 0.23608 -1.687 0.0943 .
scale(age) -0.04986 0.10228 -0.487 0.6268
gender_m 0.38275 0.24407 1.568 0.1194
gender_o -0.85183 0.46028 -1.851 0.0667 .
education_cat2_coll -0.15481 0.10712 -1.445 0.1510
urban_rural_cat2_rural -0.02897 0.09148 -0.317 0.7520
target1 0.39454 0.24553 1.607 0.1107
target2 0.13913 0.26312 0.529 0.5979
target3 0.45856 0.23478 1.953 0.0531 .
target4 -0.02524 0.24092 -0.105 0.9167
target5 0.05631 0.24920 0.226 0.8216
target6 -0.12212 0.23673 -0.516 0.6069
target7 -0.97545 0.23967 -4.070 8.42e-05 ***
target8 0.36409 0.24140 1.508 0.1341
target9 -0.11703 0.27463 -0.426 0.6708
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9355 on 121 degrees of freedom
(14 observations deleted due to missingness)
Multiple R-squared: 0.2099, Adjusted R-squared: 0.1185
F-statistic: 2.296 on 14 and 121 DF, p-value: 0.007933Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + scale(education_catX) + urban_rural_cat2 +
+(1 | target)
Data: d_sim_us_th %>% mutate(education_catX = as.numeric(education_catX))
REML criterion at convergence: 385
Scaled residuals:
Min 1Q Median 3Q Max
-2.41827 -0.67531 0.09798 0.79061 1.58834
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.1140 0.3376
Residual 0.8686 0.9320
Number of obs: 136, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) -0.43240 0.25506 73.12575 -1.695 0.0943 .
scale(age) -0.04663 0.09722 124.49344 -0.480 0.6323
gender_m 0.36711 0.24108 124.81442 1.523 0.1303
gender_o -0.85408 0.45232 126.46275 -1.888 0.0613 .
scale(education_catX) -0.16103 0.09799 126.58868 -1.643 0.1028
urban_rural_cat2_rural -0.02029 0.08875 124.48547 -0.229 0.8195
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Correlation of Fixed Effects:
(Intr) scl(g) gndr_m gendr_ sc(_X)
scale(age) 0.012
gender_m -0.770 -0.041
gender_o 0.843 0.008 -0.934
scl(dctn_X) -0.022 0.510 0.001 -0.012
urbn_rrl_2_ -0.084 -0.073 0.003 0.032 0.145Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + education_cat2 + urban_rural_cat2 +
+(1 | target)
Data: d_sim_us_th
REML criterion at convergence: 385.4
Scaled residuals:
Min 1Q Median 3Q Max
-2.37876 -0.64097 0.08029 0.78479 1.55234
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.1104 0.3323
Residual 0.8737 0.9347
Number of obs: 136, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) -0.41713 0.25542 74.51282 -1.633 0.1067
scale(age) -0.04813 0.10121 125.34264 -0.476 0.6352
gender_m 0.37575 0.24180 124.85578 1.554 0.1227
gender_o -0.88541 0.45370 126.47088 -1.952 0.0532 .
education_cat2_coll -0.15386 0.10501 128.01323 -1.465 0.1453
urban_rural_cat2_rural -0.03009 0.09054 125.28634 -0.332 0.7402
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Correlation of Fixed Effects:
(Intr) scl(g) gndr_m gendr_ edc_2_
scale(age) -0.015
gender_m -0.769 -0.053
gender_o 0.841 0.032 -0.934
edctn_ct2_c -0.067 0.560 -0.024 0.033
urbn_rrl_2_ -0.095 -0.008 -0.003 0.041 0.233
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + scale(education_catX) + urban_rural_cat2 +
target, data = d_sim_us_th %>% mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.5198 -0.5897 0.1544 0.8069 1.7317
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.96827 0.18469 -5.243 1.58e-07 ***
scale(age) -0.04712 0.06597 -0.714 0.4751
gender_m 0.28332 0.18933 1.496 0.1345
gender_o -0.58308 0.36233 -1.609 0.1076
scale(education_catX) -0.11774 0.06739 -1.747 0.0806 .
urban_rural_cat2_rural -0.03029 0.06050 -0.501 0.6166
target1 0.24592 0.16082 1.529 0.1262
target2 0.05878 0.17230 0.341 0.7330
target3 0.34132 0.15339 2.225 0.0261 *
target4 0.02851 0.16243 0.175 0.8607
target5 0.07175 0.16613 0.432 0.6658
target6 -0.02791 0.16013 -0.174 0.8616
target7 -0.71032 0.17943 -3.959 7.53e-05 ***
target8 0.25889 0.15853 1.633 0.1025
target9 -0.13803 0.18780 -0.735 0.4624
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 10.070 1.174 8.575 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 78.85 on 16 Df
Pseudo R-squared: 0.2113
Number of iterations: 25 (BFGS) + 2 (Fisher scoring)
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + education_cat2 + urban_rural_cat2 +
target, data = d_sim_us_th %>% mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.4810 -0.5877 0.2059 0.8117 1.7155
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.95663 0.18517 -5.166 2.39e-07 ***
scale(age) -0.05056 0.06862 -0.737 0.4612
gender_m 0.28897 0.18950 1.525 0.1273
gender_o -0.60678 0.36264 -1.673 0.0943 .
education_cat2_coll -0.11594 0.07221 -1.606 0.1083
urban_rural_cat2_rural -0.03854 0.06184 -0.623 0.5331
target1 0.25454 0.16207 1.571 0.1163
target2 0.04178 0.17512 0.239 0.8114
target3 0.33357 0.15353 2.173 0.0298 *
target4 0.01577 0.16203 0.097 0.9225
target5 0.07896 0.16626 0.475 0.6348
target6 -0.02272 0.16038 -0.142 0.8873
target7 -0.70516 0.17996 -3.918 8.91e-05 ***
target8 0.25549 0.15877 1.609 0.1076
target9 -0.12780 0.18777 -0.681 0.4961
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 10.03 1.17 8.576 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 78.59 on 16 Df
Pseudo R-squared: 0.2078
Number of iterations: 25 (BFGS) + 2 (Fisher scoring) China
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_us_ch)
Residuals:
Min 1Q Median 3Q Max
-1.6401 -0.9270 0.2186 0.9429 1.4894
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -8.630e-17 8.582e-02 0.000 1.000
scale(age) -7.607e-02 8.614e-02 -0.883 0.379
Residual standard error: 1.001 on 134 degrees of freedom
Multiple R-squared: 0.005787, Adjusted R-squared: -0.001633
F-statistic: 0.7799 on 1 and 134 DF, p-value: 0.3788
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_us_ch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.9881 -0.7190 0.3622 0.8740 1.2973
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.76385 0.06789 -11.251 <2e-16 ***
scale(age) -0.06363 0.06659 -0.956 0.339
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 6.260 0.716 8.743 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 54.15 on 3 Df
Pseudo R-squared: 0.007092
Number of iterations: 14 (BFGS) + 1 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_us_ch)
Residuals:
Min 1Q Median 3Q Max
-1.6141 -0.9148 0.1855 0.9476 1.4303
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.01251 0.08633 0.145 0.885
gender_m 0.06822 0.08633 0.790 0.431
Residual standard error: 1.001 on 133 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.004673, Adjusted R-squared: -0.00281
F-statistic: 0.6245 on 1 and 133 DF, p-value: 0.4308
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_us_ch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.9488 -0.7116 0.3493 0.8853 1.2490
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.75439 0.06819 -11.064 <2e-16 ***
gender_m 0.06240 0.06649 0.938 0.348
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 6.2494 0.7172 8.714 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 53.45 on 3 Df
Pseudo R-squared: 0.006282
Number of iterations: 13 (BFGS) + 1 (Fisher scoring) Race/ethnicity
Education
Call:
lm(formula = scale(MSE) ~ scale(education_catX), data = d_sim_us_ch %>%
mutate(education_catX = as.numeric(education_catX)))
Residuals:
Min 1Q Median 3Q Max
-1.5596 -0.8794 0.1408 0.9548 1.3884
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.000251 0.086838 -0.003 0.998
scale(education_catX) -0.018471 0.087164 -0.212 0.832
Residual standard error: 1.005 on 132 degrees of freedom
(2 observations deleted due to missingness)
Multiple R-squared: 0.0003401, Adjusted R-squared: -0.007233
F-statistic: 0.04491 on 1 and 132 DF, p-value: 0.8325
Call:
betareg(formula = MSE_rescaled ~ scale(education_catX), data = d_sim_us_ch %>% mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.8521 -0.6530 0.3050 0.8965 1.2015
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.76389 0.06871 -11.118 <2e-16 ***
scale(education_catX) -0.01453 0.06715 -0.216 0.829
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 6.1868 0.7124 8.685 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 52.7 on 3 Df
Pseudo R-squared: 0.0003308
Number of iterations: 12 (BFGS) + 2 (Fisher scoring) 
Call:
lm(formula = scale(MSE) ~ education_cat2, data = d_sim_us_ch)
Residuals:
Min 1Q Median 3Q Max
-1.5988 -0.8369 0.1694 0.9355 1.4230
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.007521 0.087503 0.086 0.932
education_cat2_coll -0.057860 0.087503 -0.661 0.510
Residual standard error: 1.004 on 132 degrees of freedom
(2 observations deleted due to missingness)
Multiple R-squared: 0.003301, Adjusted R-squared: -0.004249
F-statistic: 0.4372 on 1 and 132 DF, p-value: 0.5096
Call:
betareg(formula = MSE_rescaled ~ education_cat2, data = d_sim_us_ch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.9182 -0.5995 0.3323 0.8721 1.2363
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.75733 0.06907 -10.965 <2e-16 ***
education_cat2_coll -0.05171 0.06735 -0.768 0.443
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 6.2123 0.7155 8.682 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 52.97 on 3 Df
Pseudo R-squared: 0.004247
Number of iterations: 13 (BFGS) + 2 (Fisher scoring) Rural/urban
Call:
lm(formula = scale(MSE) ~ urban_rural_cat2, data = d_sim_us_ch)
Residuals:
Min 1Q Median 3Q Max
-1.6540 -1.0079 0.1715 0.8804 1.4564
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.01616 0.08672 -0.186 0.852
urban_rural_cat2_rural -0.13678 0.08672 -1.577 0.117
Residual standard error: 0.9975 on 133 degrees of freedom
(1 observation deleted due to missingness)
Multiple R-squared: 0.01836, Adjusted R-squared: 0.01098
F-statistic: 2.488 on 1 and 133 DF, p-value: 0.1171
Call:
betareg(formula = MSE_rescaled ~ urban_rural_cat2, data = d_sim_us_ch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.9699 -0.7943 0.3373 0.8410 1.2550
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.77628 0.06903 -11.24 <2e-16 ***
urban_rural_cat2_rural -0.09550 0.06727 -1.42 0.156
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 6.2690 0.7198 8.71 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 53.9 on 3 Df
Pseudo R-squared: 0.01448
Number of iterations: 14 (BFGS) + 2 (Fisher scoring) Religion
Call:
lm(formula = scale(MSE) ~ religion_cat3, data = d_sim_us_ch)
Residuals:
Min 1Q Median 3Q Max
-1.7062 -0.8712 0.1571 0.9001 1.3836
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.06732 0.13503 0.499 0.619
religion_cat3_buddhist -0.07633 0.18454 -0.414 0.680
religion_cat3_other 0.10542 0.23106 0.456 0.649
Residual standard error: 0.9743 on 111 degrees of freedom
(22 observations deleted due to missingness)
Multiple R-squared: 0.001959, Adjusted R-squared: -0.01602
F-statistic: 0.1089 on 2 and 111 DF, p-value: 0.8969
Call:
betareg(formula = MSE_rescaled ~ religion_cat3, data = d_sim_us_ch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.2177 -0.6896 0.3119 0.8983 1.2332
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.71150 0.10376 -6.857 7.02e-12 ***
religion_cat3_buddhist -0.02217 0.14076 -0.158 0.875
religion_cat3_other 0.04743 0.17554 0.270 0.787
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 6.7684 0.8474 7.987 1.38e-15 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 47.34 on 4 Df
Pseudo R-squared: 0.0006548
Number of iterations: 15 (BFGS) + 2 (Fisher scoring) Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_us_ch, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-1.6974 -0.8158 0.1577 0.7392 2.0600
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.001253 0.082341 -0.015 0.98788
target1 -0.464532 0.243725 -1.906 0.05893 .
target2 0.165268 0.243725 0.678 0.49896
target3 0.387510 0.243725 1.590 0.11435
target4 0.464226 0.251892 1.843 0.06769 .
target5 -0.298024 0.243725 -1.223 0.22369
target6 -0.053026 0.243725 -0.218 0.82812
target7 -0.718161 0.251892 -2.851 0.00509 **
target8 0.048520 0.251892 0.193 0.84756
target9 0.433270 0.243725 1.778 0.07786 .
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9596 on 126 degrees of freedom
Multiple R-squared: 0.1405, Adjusted R-squared: 0.07912
F-statistic: 2.289 on 9 and 126 DF, p-value: 0.02056
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_us_ch)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.2183 -0.8881 0.2734 0.7980 1.7910
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.77881 0.06425 -12.121 < 2e-16 ***
target1 -0.39628 0.19423 -2.040 0.04132 *
target2 0.08056 0.18291 0.440 0.65961
target3 0.33391 0.17918 1.864 0.06238 .
target4 0.39364 0.18446 2.134 0.03284 *
target5 -0.23870 0.18993 -1.257 0.20882
target6 -0.01486 0.18475 -0.080 0.93591
target7 -0.55420 0.20576 -2.693 0.00707 **
target8 0.02459 0.19011 0.129 0.89708
target9 0.37492 0.17874 2.098 0.03594 *
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 7.3215 0.8455 8.659 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 64.59 on 11 Df
Pseudo R-squared: 0.146
Number of iterations: 22 (BFGS) + 1 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + scale(education_catX) +
religion_cat3 + urban_rural_cat2 + target, data = d_sim_us_ch %>%
mutate(education_catX = as.numeric(education_catX)))
Residuals:
Min 1Q Median 3Q Max
-1.7855 -0.8826 0.2065 0.7311 1.6315
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.028043 0.146589 0.191 0.849
scale(age) 0.035009 0.115519 0.303 0.763
gender_m -0.036314 0.101044 -0.359 0.720
scale(education_catX) -0.015559 0.115801 -0.134 0.893
religion_cat3_buddhist 0.008771 0.196799 0.045 0.965
religion_cat3_other 0.002453 0.250625 0.010 0.992
urban_rural_cat2_rural -0.153559 0.106015 -1.448 0.151
target1 -0.408868 0.275045 -1.487 0.140
target2 -0.064240 0.333506 -0.193 0.848
target3 0.301786 0.257671 1.171 0.244
target4 0.431153 0.283818 1.519 0.132
target5 -0.249005 0.290031 -0.859 0.393
target6 -0.074439 0.260780 -0.285 0.776
target7 -0.438204 0.313731 -1.397 0.166
target8 0.007805 0.270807 0.029 0.977
target9 0.463229 0.291097 1.591 0.115
Residual standard error: 0.978 on 95 degrees of freedom
(25 observations deleted due to missingness)
Multiple R-squared: 0.1156, Adjusted R-squared: -0.024
F-statistic: 0.8281 on 15 and 95 DF, p-value: 0.6444
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + education_cat2 +
religion_cat3 + urban_rural_cat2 + target, data = d_sim_us_ch)
Residuals:
Min 1Q Median 3Q Max
-1.8059 -0.8842 0.1773 0.7210 1.6292
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.02903 0.14333 0.203 0.840
scale(age) 0.02406 0.11500 0.209 0.835
gender_m -0.03707 0.10088 -0.367 0.714
education_cat2_coll -0.04142 0.11263 -0.368 0.714
religion_cat3_buddhist 0.01295 0.19615 0.066 0.947
religion_cat3_other -0.00583 0.24813 -0.023 0.981
urban_rural_cat2_rural -0.15952 0.10522 -1.516 0.133
target1 -0.40541 0.27443 -1.477 0.143
target2 -0.06736 0.33199 -0.203 0.840
target3 0.29438 0.25825 1.140 0.257
target4 0.43606 0.28401 1.535 0.128
target5 -0.25177 0.28941 -0.870 0.387
target6 -0.07225 0.25994 -0.278 0.782
target7 -0.43983 0.31339 -1.403 0.164
target8 0.01196 0.26895 0.044 0.965
target9 0.45775 0.29120 1.572 0.119
Residual standard error: 0.9774 on 95 degrees of freedom
(25 observations deleted due to missingness)
Multiple R-squared: 0.1167, Adjusted R-squared: -0.02274
F-statistic: 0.837 on 15 and 95 DF, p-value: 0.6348Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + scale(education_catX) + religion_cat3 +
urban_rural_cat2 + +(1 | target)
Data: d_sim_us_ch %>% mutate(education_catX = as.numeric(education_catX))
REML criterion at convergence: 321.6
Scaled residuals:
Min 1Q Median 3Q Max
-1.8142 -0.7211 0.1416 0.8107 1.3285
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.0143 0.1196
Residual 0.9543 0.9769
Number of obs: 111, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 0.04948 0.14606 28.34635 0.339 0.737
scale(age) 0.01140 0.11235 102.08469 0.101 0.919
gender_m 0.01161 0.09482 101.28006 0.122 0.903
scale(education_catX) -0.02298 0.11315 101.62214 -0.203 0.839
religion_cat3_buddhist -0.01219 0.19375 99.54632 -0.063 0.950
religion_cat3_other 0.03500 0.24076 103.80214 0.145 0.885
urban_rural_cat2_rural -0.14400 0.10106 103.99762 -1.425 0.157
Correlation of Fixed Effects:
(Intr) scl(g) gndr_m sc(_X) rlgn_ct3_b rlgn_ct3_t
scale(age) 0.063
gender_m 0.032 0.114
scl(dctn_X) 0.179 0.459 -0.008
rlgn_ct3_bd -0.104 -0.194 -0.020 -0.114
rlgn_ct3_th 0.533 0.150 0.018 0.186 -0.772
urbn_rrl_2_ 0.175 -0.088 -0.083 0.253 -0.113 0.166 Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + education_cat2 + religion_cat3 +
urban_rural_cat2 + +(1 | target)
Data: d_sim_us_ch
REML criterion at convergence: 321.5
Scaled residuals:
Min 1Q Median 3Q Max
-1.8524 -0.6818 0.1489 0.8165 1.3586
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.0131 0.1144
Residual 0.9535 0.9765
Number of obs: 111, groups: target, 10
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 0.052293 0.143287 27.779170 0.365 0.718
scale(age) -0.001931 0.111189 102.952528 -0.017 0.986
gender_m 0.010737 0.094737 101.292340 0.113 0.910
education_cat2_coll -0.053152 0.110796 100.336817 -0.480 0.632
religion_cat3_buddhist -0.008157 0.193233 99.409248 -0.042 0.966
religion_cat3_other 0.027370 0.239003 103.643618 0.115 0.909
urban_rural_cat2_rural -0.150318 0.100632 103.949083 -1.494 0.138
Correlation of Fixed Effects:
(Intr) scl(g) gndr_m edc_2_ rlgn_ct3_b rlgn_ct3_t
scale(age) -0.002
gender_m 0.035 0.130
edctn_ct2_c 0.038 0.442 0.027
rlgn_ct3_bd -0.089 -0.184 -0.024 -0.094
rlgn_ct3_th 0.518 0.131 0.024 0.147 -0.771
urbn_rrl_2_ 0.142 -0.101 -0.075 0.240 -0.107 0.155
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + scale(education_catX) + religion_cat3 +
urban_rural_cat2 + target, data = d_sim_us_ch %>% mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.5453 -0.8965 0.3746 0.8230 1.7245
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.758869 0.107908 -7.033 2.03e-12 ***
scale(age) 0.041071 0.083932 0.489 0.6246
gender_m -0.013005 0.073502 -0.177 0.8596
scale(education_catX) -0.002331 0.084566 -0.028 0.9780
religion_cat3_buddhist 0.029752 0.142956 0.208 0.8351
religion_cat3_other -0.029668 0.182169 -0.163 0.8706
urban_rural_cat2_rural -0.106610 0.077490 -1.376 0.1689
target1 -0.361979 0.208487 -1.736 0.0825 .
target2 -0.078838 0.243428 -0.324 0.7460
target3 0.264616 0.182843 1.447 0.1478
target4 0.354677 0.200512 1.769 0.0769 .
target5 -0.160546 0.213606 -0.752 0.4523
target6 -0.038818 0.190519 -0.204 0.8385
target7 -0.309536 0.237075 -1.306 0.1917
target8 -0.011186 0.197162 -0.057 0.9548
target9 0.380669 0.206562 1.843 0.0653 .
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 7.6468 0.9771 7.826 5.04e-15 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 52.62 on 17 Df
Pseudo R-squared: 0.1129
Number of iterations: 28 (BFGS) + 2 (Fisher scoring)
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + education_cat2 + religion_cat3 +
urban_rural_cat2 + target, data = d_sim_us_ch %>% mutate(education_catX = as.numeric(education_catX)))
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.5479 -0.8757 0.3207 0.8299 1.7071
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.759947 0.105693 -7.190 6.47e-13 ***
scale(age) 0.028921 0.083692 0.346 0.7297
gender_m -0.013364 0.073422 -0.182 0.8556
education_cat2_coll -0.029635 0.082116 -0.361 0.7182
religion_cat3_buddhist 0.034925 0.142486 0.245 0.8064
religion_cat3_other -0.039434 0.180433 -0.219 0.8270
urban_rural_cat2_rural -0.112401 0.077024 -1.459 0.1445
target1 -0.357477 0.208019 -1.718 0.0857 .
target2 -0.082636 0.242370 -0.341 0.7331
target3 0.257171 0.183360 1.403 0.1608
target4 0.357698 0.200704 1.782 0.0747 .
target5 -0.161388 0.213190 -0.757 0.4490
target6 -0.039133 0.189964 -0.206 0.8368
target7 -0.309594 0.236850 -1.307 0.1912
target8 -0.006849 0.195916 -0.035 0.9721
target9 0.376658 0.206667 1.823 0.0684 .
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 7.6547 0.9782 7.826 5.05e-15 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 52.69 on 17 Df
Pseudo R-squared: 0.1145
Number of iterations: 29 (BFGS) + 2 (Fisher scoring) Vanuatu
Age
Call:
lm(formula = scale(MSE) ~ scale(age), data = d_sim_us_vt)
Residuals:
Min 1Q Median 3Q Max
-1.3050 -1.0483 -0.1352 1.0298 1.4923
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.001327 0.084256 0.016 0.987
scale(age) -0.035285 0.084554 -0.417 0.677
Residual standard error: 1.004 on 140 degrees of freedom
(6 observations deleted due to missingness)
Multiple R-squared: 0.001242, Adjusted R-squared: -0.005892
F-statistic: 0.1741 on 1 and 140 DF, p-value: 0.6771
Call:
betareg(formula = MSE_rescaled ~ scale(age), data = d_sim_us_vt)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.6437 -1.0030 0.1053 0.9601 1.2703
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.92336 0.07184 -12.853 <2e-16 ***
scale(age) -0.02107 0.06920 -0.304 0.761
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 5.5929 0.6277 8.91 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 59.1 on 3 Df
Pseudo R-squared: 0.0007607
Number of iterations: 13 (BFGS) + 1 (Fisher scoring) Gender
Call:
lm(formula = scale(MSE) ~ gender, data = d_sim_us_vt)
Residuals:
Min 1Q Median 3Q Max
-1.337 -1.047 -0.132 1.015 1.490
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.003007 0.089103 0.034 0.973
gender_m 0.007947 0.089103 0.089 0.929
Residual standard error: 1.003 on 146 degrees of freedom
Multiple R-squared: 5.448e-05, Adjusted R-squared: -0.006794
F-statistic: 0.007954 on 1 and 146 DF, p-value: 0.9291
Call:
betareg(formula = MSE_rescaled ~ gender, data = d_sim_us_vt)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-1.6613 -0.9914 0.1019 0.9484 1.2625
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.926602 0.075681 -12.244 <2e-16 ***
gender_m -0.004612 0.072886 -0.063 0.95
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 5.5831 0.6138 9.096 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 61.59 on 3 Df
Pseudo R-squared: 2.835e-05
Number of iterations: 13 (BFGS) + 2 (Fisher scoring) Location
Call:
lm(formula = scale(MSE) ~ location_cat2, data = d_sim_us_vt)
Residuals:
Min 1Q Median 3Q Max
-1.5562 -1.0115 -0.0272 0.8939 1.7147
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -3.650e-17 8.026e-02 0.000 1.00000
location_cat2_urban 2.298e-01 8.026e-02 2.863 0.00481 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9764 on 146 degrees of freedom
Multiple R-squared: 0.05317, Adjusted R-squared: 0.04668
F-statistic: 8.198 on 1 and 146 DF, p-value: 0.00481
Call:
betareg(formula = MSE_rescaled ~ location_cat2, data = d_sim_us_vt)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-2.0087 -1.1682 0.1931 0.8879 1.4244
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.92996 0.06943 -13.393 <2e-16 ***
location_cat2_urban 0.16632 0.06666 2.495 0.0126 *
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 5.8260 0.6423 9.07 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 64.65 on 3 Df
Pseudo R-squared: 0.04225
Number of iterations: 13 (BFGS) + 2 (Fisher scoring) Target
Call:
lm(formula = scale(MSE) ~ target, data = d_sim_us_vt, contrasts = list(target = "contr.sum"))
Residuals:
Min 1Q Median 3Q Max
-2.1635 -0.7050 -0.1090 0.7198 1.8083
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.03220 0.08829 0.365 0.7159
target1 -0.48900 0.22617 -2.162 0.0323 *
target2 0.28486 0.22617 1.260 0.2100
target3 0.12905 0.24047 0.537 0.5924
target4 0.31075 0.57705 0.539 0.5911
target5 0.12944 0.23292 0.556 0.5793
target6 -0.14580 0.22617 -0.645 0.5202
target7 -0.31136 0.22617 -1.377 0.1709
target8 -1.08023 0.22617 -4.776 4.54e-06 ***
target9 0.33796 0.23292 1.451 0.1491
target10 -0.05567 0.22617 -0.246 0.8059
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8916 on 137 degrees of freedom
Multiple R-squared: 0.2592, Adjusted R-squared: 0.2051
F-statistic: 4.793 on 10 and 137 DF, p-value: 6.744e-06
Call:
betareg(formula = MSE_rescaled ~ target, data = d_sim_us_vt)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.1002 -0.6754 0.0639 0.8512 1.7259
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.91934 0.07430 -12.374 < 2e-16 ***
target1 -0.45633 0.19901 -2.293 0.0218 *
target2 0.16604 0.18352 0.905 0.3656
target3 0.06323 0.19725 0.321 0.7485
target4 0.37066 0.45753 0.810 0.4179
target5 0.09951 0.19029 0.523 0.6010
target6 -0.08487 0.18883 -0.449 0.6531
target7 -0.18554 0.19134 -0.970 0.3322
target8 -0.92474 0.21434 -4.314 1.6e-05 ***
target9 0.26936 0.18714 1.439 0.1500
target10 -0.01217 0.18714 -0.065 0.9482
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 7.3600 0.8227 8.947 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 81.51 on 12 Df
Pseudo R-squared: 0.2574
Number of iterations: 21 (BFGS) + 1 (Fisher scoring) All together
Call:
lm(formula = scale(MSE) ~ scale(age) + gender + location_cat2 +
target, data = d_sim_us_vt)
Residuals:
Min 1Q Median 3Q Max
-2.37496 -0.57032 -0.03288 0.65019 2.01186
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.06735 0.09407 0.716 0.47531
scale(age) -0.05112 0.07494 -0.682 0.49637
gender_m 0.02968 0.08134 0.365 0.71576
location_cat2_urban 0.20932 0.07438 2.814 0.00567 **
target1 -0.53369 0.22410 -2.381 0.01872 *
target2 0.33815 0.23033 1.468 0.14452
target3 0.08882 0.23698 0.375 0.70844
target4 0.56692 0.57601 0.984 0.32687
target5 0.09823 0.22937 0.428 0.66918
target6 -0.26657 0.23636 -1.128 0.26151
target7 -0.31378 0.22388 -1.402 0.16347
target8 -1.07607 0.23039 -4.671 7.48e-06 ***
target9 0.34600 0.22895 1.511 0.13319
target10 -0.09678 0.23167 -0.418 0.67684
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8739 on 128 degrees of freedom
(6 observations deleted due to missingness)
Multiple R-squared: 0.3082, Adjusted R-squared: 0.2379
F-statistic: 4.386 on 13 and 128 DF, p-value: 4.248e-06Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
Formula: scale(MSE) ~ scale(age) + gender + location_cat2 + +(1 | target)
Data: d_sim_us_vt
REML criterion at convergence: 389
Scaled residuals:
Min 1Q Median 3Q Max
-2.51911 -0.74027 -0.02537 0.83385 2.19707
Random effects:
Groups Name Variance Std.Dev.
target (Intercept) 0.2248 0.4741
Residual 0.7624 0.8731
Number of obs: 142, groups: target, 11
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 0.03352 0.16620 10.50053 0.202 0.84400
scale(age) -0.04136 0.07444 130.20734 -0.556 0.57940
gender_m 0.02055 0.08058 131.89639 0.255 0.79906
location_cat2_urban 0.20776 0.07397 129.87567 2.809 0.00575 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Correlation of Fixed Effects:
(Intr) scl(g) gndr_m
scale(age) -0.024
gender_m 0.181 -0.065
lctn_ct2_rb 0.013 0.067 -0.032
Call:
betareg(formula = MSE_rescaled ~ scale(age) + gender + location_cat2 + target, data = d_sim_us_vt)
Standardized weighted residuals 2:
Min 1Q Median 3Q Max
-3.5086 -0.5677 0.0677 0.8122 1.8919
Coefficients (mean model with logit link):
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.89618 0.07848 -11.420 < 2e-16 ***
scale(age) -0.03253 0.06276 -0.518 0.60423
gender_m 0.02976 0.06773 0.439 0.66034
location_cat2_urban 0.16971 0.06215 2.731 0.00632 **
target1 -0.51796 0.19745 -2.623 0.00871 **
target2 0.23587 0.18440 1.279 0.20086
target3 0.03755 0.19377 0.194 0.84633
target4 0.57194 0.45596 1.254 0.20972
target5 0.08506 0.18665 0.456 0.64859
target6 -0.19547 0.20002 -0.977 0.32845
target7 -0.18048 0.18889 -0.955 0.33934
target8 -0.94714 0.21899 -4.325 1.53e-05 ***
target9 0.29123 0.18322 1.589 0.11195
target10 -0.04096 0.19104 -0.214 0.83022
Phi coefficients (precision model with identity link):
Estimate Std. Error z value Pr(>|z|)
(phi) 7.8236 0.8955 8.737 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Type of estimator: ML (maximum likelihood)
Log-likelihood: 82.17 on 15 Df
Pseudo R-squared: 0.2949
Number of iterations: 23 (BFGS) + 2 (Fisher scoring) Summary stats
Group-level average matching scores (observed)
Theoretical range for MSE
Kinda = 0.5
Kinda = yes = 1
MSE summary stats by group (observed)
Column `education_cat2` joining factors with different levels, coercing to character vectorColumn `ethnicity_cat2` joining factors with different levels, coercing to character vectorColumn `gender` joining factors with different levels, coercing to character vectorColumn `education_cat2` joining character vector and factor, coercing into character vectorColumn `ethnicity_cat2` joining character vector and factor, coercing into character vectorColumn `religion_cat3` joining factors with different levels, coercing to character vectorColumn `target` joining factors with different levels, coercing to character vectorColumn `gender` joining character vector and factor, coercing into character vectorColumn `education_cat2` joining character vector and factor, coercing into character vectorColumn `ethnicity_cat2` joining character vector and factor, coercing into character vectorColumn `religion_cat3` joining character vector and factor, coercing into character vectorColumn `target` joining character vector and factor, coercing into character vectorColumn `gender` joining character vector and factor, coercing into character vectorColumn `target` joining character vector and factor, coercing into character vectorColumn `gender` joining character vector and factor, coercing into character vectorColumn `ethnicity_cat2` joining character vector and factor, coercing into character vectorColumn `religion_cat3` joining character vector and factor, coercing into character vectorColumn `target` joining character vector and factor, coercing into character vectorColumn `gender` joining character vector and factor, coercing into character vectorColumn `religion_cat3` joining character vector and factor, coercing into character vectorColumn `target` joining character vector and factor, coercing into character vectorColumn `gender` joining character vector and factor, coercing into character vectorColumn `religion_cat3` joining character vector and factor, coercing into character vectorColumn `target` joining character vector and factor, coercing into character vectorColumn `gender` joining character vector and factor, coercing into character vectorColumn `target` joining character vector and factor, coercing into character vectorColumn `gender` joining character vector and factor, coercing into character vectorColumn `location_cat2` has different attributes on LHS and RHS of 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